Evaluation of some alternatives for economic development of extractive activities in Acre, Brazil

( q~

EVALUATION OF SOME ALTERNATIVES FOR
ECONOMIC DEVELOPMENT OF
EXTRACTIVE ACTIVITIES IN ACRE, BRAZIL

By
HELIODORO ARGUELLO ARIAS

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA

1996

EVALUATION OF SOME ALTERNATIVES FOR
ECONOMIC DEVELOPMENT OF
EXTRACTIVE ACTIVITIES IN ACRE, BRAZIL

By

HELIODORO ARGUELLO ARIAS

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

1996

I dedicate this dissertation to my wife, Isabel, my son,

Christian, and my daughter, Karen, who have always been close

to me and served as my inspiration. Additionally, I extend my

dedication to the rubber tapper families in the Antimari State

Forest, hoping that this research helps them.

ACKNOWLEDGMENTS

This study was conducted with the help and support of

several people and organizations. I wish to thank the

members of my supervisory committee, Dr. Cesar Caviedes,

chairman, Dr. Nigel Smith, co-chairman, and Drs. Marianne

Schmink, Mary Duryea, Peter Hildebrand, and Abe Goldman, for

their guidance and counsel during my course work and the

writing stages of this research.

The following people from the University of Florida

also contributed to this research: Dr. Steven Sanderson, who

gave me the initial opportunity to come to Gainesville with

the support of the Tropical Conservation and Development

(TCD) Program; Dr. Ben Blair, who provided helpful advice

during the planning phase of this research; and Dr. Francis

Putz, who suggested relevant literature on forest

management.

I also would like to thank the members of PESACRE and

FUNTAC in Rio Branco, who provided support and interest in

this research. Special appreciation goes to the Co-Director

of the TCD Program in the Center for Latin American Studies,

Dr. Marianne Schmink, for her confidence, guidance, and

support in all phases of this study.

iii

I want to thank especially the people of the Antimari

State Forest in Acre, Brazil, for allowing me to disturb

their schedules and privacy. Likewise, I would like to

acknowledge the people who helped me conduct the market

studies in eleven Brazilian cities. Special thanks go to

the Salviano and Magnavita families in S&o Paulo and Salvador

for their great hospitality.

Funding for this research was provided by the TCD

Program of the Center for Latin American Studies at the

University of Florida.

TABLE OF CONTENTS

ACKNOWLEDGMENTS . . . .. . ii

LIST OF TABLES . . . .... ix

LIST OF FIGURES . . . .... .xii

ABSTRACT . . . .. . xiii

CHAPTER 1 INTRODUCTION . . . . 1

CHAPTER 2 A REVIEW OF THE ROLE OF EXTRACTIVE ACTIVITIES
IN THE AMAZON AND THE ALTERNATIVES FOR
IMPROVEMENT . . . . 5
Introduction . . . . ... 5
Economic Settlement of the Amazon and Extractive
Activities . . .. 6
The Conquest . . . . 7
The Occupation . . . 7
The Rubber Cycle . . . 8
The Valuation . . . .13
The Integration . . .. .. 14
Extractivism As Inefficient Land Use Practice 16
Alternatives for Improving Extractive Activities 30
Enriching the Forest by Planting Economically
Valuable Plants and Removing Competitors
. . . . 31
Utilizing the Natural High Densities of
Certain Tree Species Which Are Scattered
Throughout the Amazon Region . .. 32
Increasing the Value Added to Forest Products
Before They Leave the Reserves .... 35
Increasing the Number of Species That Are
Extracted . . . ... .35
Non-timber products . . .. .36
Medicinal plants . .. . 38
Timber products . .. . 40
Agroforestry . . . .. 51
Creating Income-Generating Forests in the
Amazon . . . . 58
Strategies for Enhancing Income
Generation . . ... 62
Conclusions . . . . ... 63

CHAPTER 3 STUDY SITE AND METHODOLOGY . ... 66
Location of study site . . .. . 66
Socioeconomic Aspects . . .. 68
Production Systems . . ... 71
Fauna . . . . . 73
Vegetation . . .. . 73
FUNTAC's sustained forest management for
multiple use of the ASF .. .... 81
Research design . . . ... 85

CHAPTER 4 MARKETS AND POTENTIAL MARKETS FOR FOREST
PRODUCTS ... ....... .. .. .. 90
Market for non-timber forest products (NTFPs) .. 91
Rubber and Brazil nuts ...... ... 91
Heart-of palm ..... ...... .. 97
Copaiba oil . . .. . 97
Andiroba oil . . .. . 99
Medicinal plants . . ... 99
Green market in the USA . .. . 106
Timber forest products . . ... 106
Timber Species of High, Middle, and Low
quality . . .. . 109
Green market of TFP in USA .. . 115
Conclusions . . . .. . 116

CHAPTER 5 DESCRIPTION AND ACTIVITIES OF THE ASF'S
HOUSEHOLDS . . . .. . 121
Introduction ............ .. 121
Characterization of the Households in the ASF 121
ASF Production Units Grouped by Activities and
Location . . . ... 122
Group I . . .. . 124
Group II . . .. . 124
Group III . . .. . 124
Group IV . . . . 124
Group V . . . . 124
Family Background and Calendar of Activities 125
Extractive activities .. . 128
Agricultural activities . ... .132
Animal husbandry . . . 134
Housekeeping . . ... 135
Off-farm activities . .. . 136
Allowances . . ..... .. 136
Labor Distribution by Activities and by Family
Members . . . ... 137
Conclusions . . . . ... 144

CHAPTER 6 LINEAR PROGRAMMING ANALYSIS FOR THE
PRODUCTION UNITS IN THE ASF . . ... .146
Introduction . . . . ... 146
Linear Programming . . . .. 146
Simulation of the Current Situation . 149
Alternatives for Improving Household Income in the
ASF .. . . .. . . 157
Species Considered in a Forest Use Plan for
the ASF 159
Production and management of non-timber
forest products (NTFPs) . 162
Production and management of timber
forest products (TFPs) . .. .166
Alternatives Considered . . .. .170
Simulated current situation (SCS) 171
Non-timber forest products (AS1) . 171
Timber forest products (AS2) . .. .172
Timber and non-timber forest products
(AS3) . . . ... 173
Results of the Linear Programming Analysis . 173
Impact on Traditional Extractive Activities 174
Impact on Traditional Agricultural
Activities . . . .. 176
Impact on the Amount of Labor Contributed by
Each Family Member with Respect to the
Total Labor Required .. . 177
Impact on Gross Return per Unit of Labor 183
Impact on Discretionary Income . .. 184
Resources That Limit the Optimal Solutions in
the Evaluated Alternative Solutions 185
Aggregated Quantities of Discretionary Income,
Land Cleared and Products from ASF
Activities . . . ... 188
Conclusions . . .. . . 192

CHAPTER 7 DISCUSSION AND RECOMMENDATIONS . 195
Recommendations .. ............ 198
Conditions Required for the Development of
Activities According to the Optimal
Solutions for the Different Simulations
Performed .. . . . 200
Property rights . . ... .200
Sustainable management . ... .201
Production infrastructure and
organization . . .. 202
Post-harvest processing . .. 208
Transportation and marketing . 208
Implications of Findings . . ... 208
Ecological Sustainability . .. . 209
How the timber harvesting should be
done . ... . 211
Social Sustainability . . ... 214

vii

APPENDIX A DESCRIPTION OF SPECIES WITH ECONOMIC
POTENTIAL PRESENT IN THE ASF . . .. .218

APPENDIX B LINEAR PROGRAMMING TABLES . .. .246

REFERENCE LIST . . . . ... 272

BIOGRAPHICAL SKETCH . . . . .. .285

viii

LIST OF TABLES

Table page

Table 2.1 Economic indicators of different extractive
activities in the Amazon. . . . 19
Table 2.2 Shelled and in-shell Brazil nuts, by
countries, imported by the USA between 1989 and
1993. . . . . . 23
Table 2.3 Fertilizer requirements for continuous
cultivation of three crops per year in an Ultisol
in Yurimaguas, Peru. . . .... 28
Table 2.4 Non-timber products and economic indicators of
"riberinho" activities in the Tapaj6s National
Forest, Santar6m, Brazil. . ... . 39
Table 2.5 Harvest intensity per year and area following
traditional exploration methods in Abufari,
Amazonas, Brazil. . . . . 44
Table 2.6 Natural regeneration after 10 years, following
traditional methods for timber exploration in
Abufari, Amazonas, Brazil. . . ... 45
Table 2.7 Brazilian manufactured wood-based articles
imported by the USA between 1989 and 1993. ... 48
Table 2.8 Timber products exported by region between
1961 and 1990. . . ...... 49
Table 2.9 Destination of timber products exported from
Bel6m (Para), Brazil, in 1993. . . 51
Table 2.10 Gross returns and activities in the optimal
production plans for the current and alternative
situations in the Cachoeira extractive settlement,
Acre, Brazil. . . . . 52
Table 2.11 Some Brazilian natural non-timber products
imported by the USA between 1989 and 1993. ... 55
Table 2.12 Risk analysis performed by ECOTEC on 15
selected Amazonian products. . . ... 57
Table 3.1 Ingested plant residues found in the stomachs
of about 300 animals from the ASF. . ... 74
Table 3.2 ASF management areas, based on forest types. 74
Table 3.3 Main parameters determining the commercial
potential for wood in the ASF, by management
region. . . . . ... 76
Table 3.4 Timber potential of the 82 tree species
selected according to their potential for timber,
by management region. . . . .. 77

Table 3.5 Average revenues and costs of different logged
volumes potentially harvested in the ASF. 77
Table 3.6 Log production costs, based on different
harvesting intensities in the ASF. . .. 78
Table 3.7 Potential uses of different vegetal life forms
in the ASF. . ..... . 79
Table 3.8 Wood density of 20 species selected for timber
and furniture from the ASF. . . 82
Table 4.1 Prices of some non-timber products and
quantity demanded by some retailers and wholesalers
in different Brazilian cities. . . .. 100
Table 4.2 Prices for medicinal plants marketed in the
cities of Rio Branco and Cuiab& between 1990 and
1994. . .. . . . 104
Table 4.3 Non-timber products commercialized in the
USA's green market. . . . ... 107
Table 4.4 Timber prices for the most demanded species
between 1990 and 1995 in Rio Branco, Acre. 110
Table 4.5 Prices of some timber products in different
Brazilian cities. .... ........... ill
Table 4.6 Participation (%) of sawnwood exports with
respect to the total timber exports from Beldm,
Brazil, in 1993. . . . 112
Table 4.7 Exclusive and non-exclusive "smart wood"
companies commercializing tropical timber
products. .. ........... 117
Table 5.1 Family background of the production units by
groups. ................ . 126
Table 5.2 Resources used and production levels for each
group of households located near and far from the
river in the ASF. .............. ..130
Table 5.3 Labor distribution of each active member by
activities performed in the households of Group I,
on the river banks. . . .... ..138
Table 5.4 Labor distribution of each active member by
activities performed in the households of Group II,
on the river banks. . . .. .. 139
Table 5.5 Labor distribution of each active member by
activities performed in the households of Group
III, on the river banks.... . . 140
Table 5.6 Labor distribution of each active member by
activities performed in the households of Group IV,
far from the river. . . . 141
Table 5.7 Labor distribution of each active member and
by activities performed in the households of Group
V, far from the river. .. . . 142
Table 6.1 LP model in a matrix form, involving all the
activities performed by five household groups in
the ASF, Acre, Brazil. ............. .148
Table 6.2 Comparison of main resources used and
production levels, observed and simulated for each
group of households in the ASF, Acre, Brazil. 154

Table 6.3 Comparison of labor employed in each activity,
observed and simulated for each group of households
in the ASF, Acre, Brazil. . . .. 156
Table 6.4 Relative frequencies of selected species in
the colocag6es of the ASF. . . ... 161
Table 6.5 Palms of economic interest found in the ASF,
Acre, Brazil. . . . . ... 165
Table 6.6 Sawn timber production costs per year in a
traditional selective production system in the ASF,
Acre, Brazil. ... . . . 170
Table 6.7 Economic indicators and activities included in
the optimum production plan for each location and
group in the simulated alternative situations for
the ASF . . . . . 175
Table 6.8 Labor used in the current and alternative
situations by each active member of the households
of Group I, on the river bank. . .. .178
Table 6.9 Labor used in the current and alternative
situations by each active member of the households
of Group II, on the river bank. .. . 179
Table 6.10 Labor used in the current and alternative
situations, by each active member of the households
of Group III, on the river bank... . 180
Table 6.11 Labor used in the current and alternative
situations, by each active member of the households
of Group IV, in the reserve's interior. . 181
Table 6.12 Labor used in the current and alternative
situations, by each active member of the households
of Group V, in the reserve's interior. . ... 182
Table 6.13 Shadow prices for resources that limited the
optimum solutions for each group and location in
the simulated situations. . . ... ..186
Table 6.14 Aggregated quantities of resources and
products incorporated in the optimal solution. 189

LIST OF FIGURES

Figure page

Figure 3.1 Antimari State Forest (ASF) location. 67
Figure 3.2 Management regions and households studied in
the ASF. . . . . .. 75
Figure 4.1 Rubber and Brazil nut prices (US $/Kg)
between 1992 and 1994 in Rio Branco, Acre. ... 92
Figure 5.1 Frequency distribution of age from the
population of 23 households in the ASF, Acre,
Brazil. ................. 123
Figure 6.1 Suggested wood harvesting plan for a
colocagco, in a cut cycle of 20 years. . ... 169

xii

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

EVALUATION OF SOME ALTERNATIVES FOR ECONOMIC DEVELOPMENT OF
EXTRACTIVE ACTIVITIES IN ACRE, BRAZIL

By
HELIODORO ARGUELLO ARIAS

December, 1996

Chairperson: Cesar Caviedes
Cochairperson: Nigel Smith
Major Department: Geography

Reconciling conservation with the improvement of living

conditions among communities settled in the Amazon forest has

been a challenge for the communities themselves, the Brazilian

government, and national and international NGOs. Despite the

great interest in this issue among scientists, development and

conservation agencies, and politicians, uncertainty prevails

regarding the actual deterioration of Amazon ecosystems and

the improvement in living conditions of the communities

involved.

Creation of extractive reserves in Brazil was an attempt

to resolve the issue. However, after years of

experimentation, neither forest conservation nor improvement

of the rubber tappers' living conditions have been achieved;

deforestation is increasingly accentuated within the reserves,

while the communities confront severe economic hardships.

xiii

This dissertation examines proposals to improve the

economic viability of extractive activities conducted by

families in the Antimari State Forest, Acre, Brazil. One

strategy was to diversify the product base, exploiting under-

utilized resources and developing markets for new products

extracted from the forests. Extraction of timber and non-

timber products was examined as a major additional activity.

Farming systems methodology was used, including

participatory observation of the activities in each household.

Factors limiting production were identified and a set of

alternatives drawn, including timber forest products such as

lumber, and non-timber forest products such as heart-of-palm,

essential oils, and handicrafts. Using linear programming, a

simulation of the current situation was constructed as a base

to analyze other alternatives. Sample households were sorted

into five categories according to their primary economic

activities (and income levels), as well as location within the

reserve.

Findings demonstrated that rubber extraction is not a

profitable activity and should be replaced by more valuable

products. Potential for sustainable timber extraction and

exploitation of non-timber products was demonstrated under

appropriate conditions of property, production infrastructure

and organization, sustainable management, post-harvest

processing, and transportation and marketing. Revenues from

these products could increase the gross return per unit of

labor input up to ten times the current levels.

xiv

CHAPTER 1
INTRODUCTION

The deterioration of traditional extractive economies

based on Brazil nut and rubber collection threatens the

sustainability of extractive reserves in the State of Acre,

Brazil. Threats to the viability of extractive reserves

include:

* Net returns from current extractive activities are too

low.

* The market prognosis for rubber is unfavorable.

* The current market, concentrated in a few extractive

products, is volatile.

* Extractive activities are labor-intensive and

inefficient.

* Deforestation rates are increasing as rubber tappers turn

to farming in an effort to recover lost income.

A typical individual production unit, called a

"colocag&o," involves approximately 300 hectares. Within

this area, a family gathers rubber and Brazil nuts while

engaging in small scale farming, fishing, and hunting

activities.

Because of the diminishing returns from extractivism,

rubber tappers are increasingly driven to agricultural

practices, thereby exacerbating deforestation in Acre.

Alternatively, rubber tappers and Brazil nut gatherers flock

to towns and cities, where their illiteracy and lack of job

skills prevent them from gaining access to all but menial

employment. In either case, people live in extremely poor

conditions.

If forest resources are to be maintained and human

populations meet their basic needs, the economic viability

of extractive activities must be improved. One way to

achieve this is diversification of the product base, which

entails exploring under-utilized resources and developing

markets for new items extracted from forests.

Producer communities and support organizations have

little information or guidance concerning appropriate

products needed to diversify the extractive production

system. This study focused on several potential products

that could be used as alternatives to the current production

system.

General Hypothesis

This dissertation examines the hypothesis that

alternative forest products are viable, from technical and

socioeconomic perspectives, in order to diminish the current

deforestation rates in the state of Acre, Brazil. The

success of these alternatives will require certain

institutional conditions that currently do not exist.

Objectives

This research was based on a linear programming

optimization procedure used to simulate current household

activities, supplemented by forest inventories, and a long-

term management plan. It analyzed more profitable and

sustainable conditions under which household members could

utilize the forest to diminish the current deforestation

rates in the Antimari State Forest (ASF), in the state of

Acre, Brazil.

A historical view of the main aspects concerning the

role of extractive activities in the Amazon and an

assessment of their weaknesses and strengths as alternatives

for conservation and income generation are examined in

Chapter 2. Chapter 3 describes the study site at the time

the research was conducted, as well as the methodology

utilized.

The results of the market study, compiled from

inspections of 11 Brazilian market centers, are presented in

Chapter 4. Chapter 5 outlines the general characteristics

of the ASF community and differentiation of five homogeneous

groups of families (population structure, household

activities, age and gender roles, and calendar of

activities).

In Chapter 6, different options for the improvement of

the extractive economic activities in the ASF are analyzed.

Utilizing optimization procedures, the current situation is

4

first simulated for each of the homogeneous household

groups, then alternatives for improving the situation are

described, simulated, and analyzed. Finally, Chapter 7

presents the research conclusions, discussing the

feasibility of the recommended alternatives within both

local and regional frameworks.

CHAPTER 2
A REVIEW OF THE ROLE OF EXTRACTIVE ACTIVITIES IN THE AMAZON
AND THE ALTERNATIVES FOR IMPROVEMENT

Introduction

Historically, the role of extractive activities in

tropical areas has been considered crucial in some periods

and marginal in others. Among Amazonian development

alternatives, extraction of non-timber products still plays

an important role in some areas, while extraction of timber,

agriculture, cattle ranching, and mineral exploitation are

the basis of the economy in other areas.

Although often associated with depletion of natural

resources in areas of low population density, extractivism

has coexisted harmoniously with the native forest for a long

time. However, drastic changes in the Amazon's population

stimulated by migration, economic dependency on a few

products with fragile markets, and the active incorporation

of the Amazon into national and international interests,

have imposed enormous pressure on areas with extractive

activities.

The controversy over extractivism in sparsely populated

areas involves two opposing views. First, there are those

who believe extractivism is an inefficient land use,

CHAPTER 2
A REVIEW OF THE ROLE OF EXTRACTIVE ACTIVITIES IN THE AMAZON
AND THE ALTERNATIVES FOR IMPROVEMENT

Introduction

Historically, the role of extractive activities in

tropical areas has been considered crucial in some periods

and marginal in others. Among Amazonian development

alternatives, extraction of non-timber products still plays

an important role in some areas, while extraction of timber,

agriculture, cattle ranching, and mineral exploitation are

the basis of the economy in other areas.

Although often associated with depletion of natural

resources in areas of low population density, extractivism

has coexisted harmoniously with the native forest for a long

time. However, drastic changes in the Amazon's population

stimulated by migration, economic dependency on a few

products with fragile markets, and the active incorporation

of the Amazon into national and international interests,

have imposed enormous pressure on areas with extractive

activities.

The controversy over extractivism in sparsely populated

areas involves two opposing views. First, there are those

who believe extractivism is an inefficient land use,

occupying areas with potential for intensive livestock

production or agriculture. Second, there are others who

consider extractivism an initial model to achieve

conservation of biodiversity, while preserving the Amazonian

lifestyle and achieving long-term economic sustainability.

In this chapter, major aspects concerning the role of

extractive activities will be examined. The extractive

activities that should be pursued in the future will be

identified, with the purpose of establishing linkages with

conservation and development of the Amazon. To reach the

established objective, three aspects will be considered: (1)

the historical evolution of extractivism; (2) the main

objections to extractivism as an inefficient land use

practice; and (3) the alternatives to improve extractivism

as a potentially sustainable practice.

Economic Settlement of the Amazon and Extractive Activities

Miranda Neto (1985) divides the economic settlement of

the Amazon into five phases: the conquest (1500-1750), the

occupation (1750-1850), the rubber cycle (1850-1946), the

valuation (1946-1966), and the integration (1966-today).

Episodes of boom and bust of extractive products have

occurred during economically decisive periods in the Amazon.

The Conquest

The conquest phase was characterized by the extraction

of resins, fibers, oils, and medicinal plants, and the

production of cocoa, sugar, and indigo (Indigofera

suffruticosa). This phase was dominated by the mestizo

society derived mainly from Portuguese men and indigenous

women. The economy was subordinated to the metropolis of

Lisbon with very little interaction with other Brazilian

areas.

The most important primary resource products during the

colonial exploitation period were: cocoa, Brazil nut

(Bertholletia excelsa), and fish; spices such as

salsaparrilla (Smilax officinalis) and copaiba (Copaifera

multijuga); aromatics such as panquile (Piper cavalcantei)

and tonka bean (Dipteryx odorata); narcotics such as iage

(Banisteriopsis caapi) and virola (Virola sp.); stimulants

such as guarand (Pranceana cupana) and xexua (Sterculia

sp.); and fibers such as piassava (Leopoldina piassava) and

malva (Urena lobata).

The Occupation

The occupation phase was conducted for the purpose of

stimulating agricultural enterprises and integrating the

Amazon into colonial trade as an exporter of primary

resources. The political integration of Para and Amazonas

into the rest of Brazil following its independence in 1823,

The Conquest

The conquest phase was characterized by the extraction

of resins, fibers, oils, and medicinal plants, and the

production of cocoa, sugar, and indigo (Indigofera

suffruticosa). This phase was dominated by the mestizo

society derived mainly from Portuguese men and indigenous

women. The economy was subordinated to the metropolis of

Lisbon with very little interaction with other Brazilian

areas.

The most important primary resource products during the

colonial exploitation period were: cocoa, Brazil nut

(Bertholletia excelsa), and fish; spices such as

salsaparrilla (Smilax officinalis) and copaiba (Copaifera

multijuga); aromatics such as panquile (Piper cavalcantei)

and tonka bean (Dipteryx odorata); narcotics such as iage

(Banisteriopsis caapi) and virola (Virola sp.); stimulants

such as guarand (Pranceana cupana) and xexua (Sterculia

sp.); and fibers such as piassava (Leopoldina piassava) and

malva (Urena lobata).

The Occupation

The occupation phase was conducted for the purpose of

stimulating agricultural enterprises and integrating the

Amazon into colonial trade as an exporter of primary

resources. The political integration of Para and Amazonas

into the rest of Brazil following its independence in 1823,

produced social instability and disaggregation of the

initial productive system. Thus, some activities continued

to be developed in the Amazon, while others were

discontinued or moved to other areas. Coffee cultivation

was switched to regions near the Brazilian capital, while

cacao plantations were developed in Bahia. Finally, during

the first half of the 19th century, the Amazon returned to

extractivism based on the collection of rubber.

The Rubber Cycle

According to Miranda Neto (1985), four events

characterized this phase: the creation of the Amazonas

province in 1850, the introduction of steam navigation, the

extensive exploitation of "seringais" or rubber estates, and

the arrival of immigrants from the Northeast.

These events boosted the rubber economy, since they

increased the importance of the traditional economy

characterized by both extractivism and the export business.

As a consequence of this cycle, forest occupation became

crucial. "Ouro negro" fever began, and the the Acre

territory was annexed to Brazil. Profits benefited the

exporters, while the rubber tappers grew poorer due to their

dependence on traders and other external agents. Rubber

tappers were not able to meet their needs, as food had to be

imported. The high revenues generated by rubber were

concentrated in the hands of the elite, who unfortunately

wasted the profits and failed to create a resilient and

diversified economy (Weinstein, 1983).

In spite of a high demand for rubber in the

international market and the increase in rubber prices,

Brazil, its sole producer, could not increase the supply due

to both labor and capital shortages (Dean, 1987; Barham and

Coomes, 1994). Meanwhile, northeast Brazil suffered not

only a crisis in its sugar economy, but also a

transformation of the livestock production system into a

subsistence economy. This scenario worsened after the long

dry season of 1877-80 that forced more than 65,000 people to

migrate to the Amazon (Souza, 1994).

Because rubber activities were pursued exclusively,

other agricultural and extractive enterprises were

completely abandoned. River navigation and other activities

connected to rubber transportation were developed. In 1866,

three large national navigation lines were established and

the Amazon River was opened to foreign navigation.

Subsequently, The Amazon Steam Navigation Company Limited

absorbed the existing national lines. The trade of goods

and services contributed to the growth of Belfm and Manaus

as main ports (Miranda Neto, 1985).

Capital scarcity gave rise to the "debt-peonage"

(aviamento) system, based primarily on foreign capital. In

the aviamento market system, goods were supplied on credit

at inflated prices and were paid for in extractive products,

mainly rubber. Since the rubber merchants or patrons

(owners of the rubber trails) paid low prices for extractive

products, while the prices of food, clothing, and medicine

soared, rubber tappers were almost always in debt. In this

system, the overriding objective was rubber production, and

all other subsistence activities were clearly subordinated

(Allegretti, 1979). However, a recent point of view

justifies the aviamento as logical, not really exploitative.

"Debt-merchandise relations" between tappers and traders (or

patrons) provided the most efficient way of reducing the

barriers posed by high transaction costs and risks for these

extractive activities (Barham and Coomes, 1994).

The aviamento involved a long chain of debtors.

Merchants and patrons themselves were usually also in debt

to downstream merchants, who in turn, were indebted through

marketing banks to wholesalers and exporters and ultimately

to foreign banks which supply seed money (Richards, 1993).

In the literature, two reasons are given to explain the

traditional social relations that persisted during the boom.

Both are based on the relative power of patrons and traders

versus tappers. One reason suggests that the relationship

with tappers as debt peons, serfs, or slaves was maintained

by force or coercion through threat of violence (Dominguez

and Gomez, 1990). Patrons and traders whose profits

depended on exploiting workers had little to gain by passing

on to workers the benefit of price increases or by altering

the production structure for extraction or trade according

to the wishes of a foreign purchaser.

The second explanation, offered by Barbara Weinstein

(1983), argues that tapper resistance, not coercion,

conditioned the nature and durability of social relations.

Rubber workers preferred the autonomy of wild rubber

extraction to the drudgery of wage labor and effectively

managed to resist proletarianization. Their resistance was

encouraged by two important features of the industry.

First, the extracted rubber was, by law and in practice, the

property of the tapper and not of the owner of the land and

trees. Second, the high cost of monitoring the working

tapper strictly limited the degree to which patrons could

truly control their workers or curb illicit rubber sales.

Between tappers and traders, on the other hand, a lasting

alliance developed based on interlocking self-interests--the

traders' need to control the exchange of rubber and the

tappers' preference for autonomy--that effectively

frustrated local and foreign pressures to rationalize rubber

extraction and trade.

According to Coomes and Barham (1994), the most

significant way to improve the Amazonian rubber industry

would have been to introduce large-scale rubber plantations

similar to those of the Asian model. Furthermore,

plantations were attempted and failed for environmental and

technical reasons (Dean, 1987). Weinstein's view of labor

relations offers another explanation as to why this effort

failed in the Amazon: he contends that plantation

development was blocked by the lasting tapper-trader

alliance. The dismal failures of foreign firms operating

wild rubber estates with disappointing yields are well

known. Tappers apparently worked at a slower pace or failed

to report all the rubber tapped, selling part of it to

itinerant traders (thereby benefiting both themselves and

the traders). This mutual bond of self-interest brought

both parties to a stalemate; tappers controlled production

but not exchange, and traders controlled exchange but not

production. Any effort to force tappers to work more

intensely would diminish their autonomy and provoke

resistance. Further, the low density of wild rubber trees

also guaranteed tappers a degree of autonomy they could not

enjoy in a plantation.

Complementing Weinstein's point of view, Coomes and

Barham (1994) affirm that labor scarcity, mobility, and the

high cost of monitoring tapper effort and output were

empirical features of rubber production. These features ob-

viate the classic portrayal of tappers as debt peons, serfs,

or slaves kept in these relationships by coercion.

In 1876, before the auto industry boom, rubber seeds

were taken to England and then to Ceylon and Singapore.

Subsequently, in 1900, the first rubber plantation in Asia

was under way and the Brazilian rubber monopoly came to an

13

end. By 1913, Asian production was twice as large and less

expensive than Brazilian production, forcing the Brazilian

rubber price to drop from 512 in 1911 to 100 per ton in

1913-14 (Weinstein, 1983).

Due to the Brazilian rubber crisis in the international

market, people in the Amazon reverted to extracting products

from other species such as Brazil nut, balata (Manilkara

bidentata), South American palm oil or caiau61 (Elaeis

oleifera), piassava, and babassu (Attalea phalerata). When

Japanese immigrated to the Amazon, black pepper (Piper

nigrum), jute (Corchorus capsularis), and vegetables were

incorporated into the regional agrarian economy.

In 1942, during World War II, the Japanese tried to

seize control of the Asian plantations, and the allied

countries were interested again in Amazonian rubber. The

USA and Brazil signed an agreement to rapidly increase

rubber production, but when the war was over, the USA did

not buy more rubber and a new collapse beset rubber

production in Amazonia (Weinstein, 1983).

The Valuation

The goal of the valuation (1946-1960) period was to

achieve territorial occupation of the Amazon; to build an

economically stable and progressive society; to strengthen

the Amazon economy; to initiate some kind of agriculturally-

'According to Smith et al., 1992

14

based economy in previously selected areas with satisfactory

transportation systems and natural resources; to develop

industries in the cities; to utilize the forest and mineral

resources in moderation; and to improve extractive

techniques as well as the forest workers' quality of life

(Miranda Neto, 1985). However, these goals were not

achieved:

From the late 1940s to the early 1960s,
income from rubber gathering had fallen from 8% to
5% of the regional economy of the Amazon. Even
so, extractive activities in general, including
tin ore and gold prospecting, the hunting of
pelts, and the gathering of Brazil nuts, still
occupied 60 percent of the regional labor force.
Rubber was an element key in sustaining these
other survival stratagems because it was
government funded. (Dean, 1987:131)

The Integration

Starting in 1966, the federal military government

developed policies towards to integrate the Amazon into the

rest of Brazil. These policies were based on the

consideration that the Amazon is a large isolated and

underoccupied area, lacking in national security provisions,

and, in addition, a source of products and services for the

entire country (Schmink and Wood, 1992).

The PIN's (National Integration Program) first phase

was the quick construction of the Transamazon and Cuiabd-

Santardm highways. On both sides of the highways, a narrow

area was designated for colonization. After 1970, the

government's view of the Amazon changed: no longer a problem

area, it was now perceived as a pioneer area in transition.

This point of view stimulated private and government

investments in minerals, timber, and livestock with poor

social results. As the economic border expanded, local

people faced land expropriation, which forced them to either

submit to the new enterprises, look deeper in the forest for

new land, migrate to urban areas (thus swelling the marginal

population), or try to resist expropriation (Miranda Neto,

1985; Schmink and Wood, 1992).

In the north of the Amazon region, the population

increased from 330,000 in 1872 to 7,650,000 in 1985 (14

million in the entire Amazon region) (Benchimol, 1985).

Although today 65% of the population in the northern region

is concentrated in the urban areas, many people live from

the forest and shifting agriculture. In 1980, it was

estimated that 68,000 families were involved in rubber

tapping, distributed in the following states and

territories: Amazonas (32,300), Acre (23,200), Pard (8,300),

and the remainder (4,200) in Rond8nia, Amapa, and Roraima.

Assuming an average of five members per family,

approximately 340,000 people were directly dependent on

tapping native rubber trees for their livelihood

(Allegretti, 1990).

Currently, rubber tappers are living in either

autonomous or debt-peonage economic systems. In the

autonomous system, a family's or group of families'

16

landholding ("colocagAo") produces a mix of both subsistence

and market goods (Allegretti, 1979). This mix generally

consists of one or two regionally commercialized products

such as rubber (Hevea brasiliensis) and Brazil nut, combined

with other forest products of more restricted occurrence,

such as copaiba, a medicinal resin, and sorvinha (Couma

utilis), another type of latex. Alongside extraction of

forest products, families practice a shifting cultivation

that generally includes traditional crops such as manioc;

crops that can substitute for industrialized goods, such as

tobacco and sugarcane; and both native and exotic fruit

trees. The autonomous rubber tapper also engages in

economic activities such as the gathering of native plants

(for example, fruits and heart-of-palm of the assaii" palm--

Euterpe precatoria--and medicinal plants), as well as

fishing and hunting.

Extractivism As Inefficient Land Use Practice

The relatively large holdings required by rubber tapper

families have been cited as a major weakness of extractive

land use systems. "Given the naturally dispersed

distribution of most marketable forest products in tropical

forests, extractive systems typically require relatively

large continuous tracts of forest land" (Browder, 1992:176).

In addition, Anderson (1992: 69) asserts that "if land were

distributed throughout the Brazilian Amazon according to the

rubber tappers' standard, the current population of the

region would have to be reduced." The average size of a

rubber tapper's extractive area currently varies between 200

and 1,500 hectares, depending on the region. In eastern

Amazonia, it is difficult to calculate the average size of a

caboclo2 household, since the Brazil nut stands do not

belong to a specific family but rather are common property

or are claimed by a large landholder.

Although one of the most important objectives for

promoting the extractive land use system was the improvement

of the rubber tappers' standard of living, the decline has

continued, even after the emergence of the Brazilian Council

of Rubber Tappers and the legal recognition of numerous

extractive reserves. Among the reasons cited are the

increasing subdivision of holdings and the resulting

"superexploitation" of individual rubber trails in order for

households to retain their purchasing power. To offset

progressively lower income from rubber tapping, some

households have increased the area of annual crops,

producing deforestation at higher rates than in other areas

where commercial agriculture, not extraction, prevails

during the same period (Browder, 1992).

According to some surveys done in the extractive

reserve of Cachoeira in Acre, extractive production based

mainly upon rubber and Brazil nut faces several fundamental

2Riverine people of primarily mixed European and native ancestry

problems, such as low rubber quality, lack of price

guarantees, low rubber and Brazil nut prices, lack of

infrastructure for processing, increase of pasture areas,

and economic dependency on what remains of the aviamento

system (UFAC/UF, 1989). The low return to producers is

partly the result of costly and risky transport conditions,

often in treacherous weather conditions, and the high risks

of marketing extractive products in a hyper-inflationary

economy (Padoch, 1989).

The aviamento system is still practiced in many forms

throughout the Amazon. Although its practitioners suffer

from severe deprivation, there are two factors that may make

its replacement difficult: first, its capacity to supply

other necessary products obtained from remote areas; and

second, its production arrangements that force rubber

tappers to maximize the time devoted to extractive

activities (Richards, 1993). Some studies show that gross

returns from traditional extractivism (including only rubber

and Brazil nuts) are less than those some other uses of the

Amazon (Table 2.1), which leads some authors to argue that

the extractive economies are unstable over time and not

indefinitely self-sustaining.

Table 2.1. Economic indicators to different extractive
activities in the Amazon.

Costs (US $) Gross Return (US $)
Site Products House- Labor Material Ha work- US $
hold day
(Ha)
Xapuri, Acrea Rubber, Brazil 370 2.19 0.24 2.3 4.38 10.75
nut
Sao Luis do Rubber, Brazil 300 2.7
Remansob nut
Cachoeira, Rubber, Brazil 300 4.4
Acreb nut
Typical, Rubber, Brazil 220 2.26 2.01 2.5 3.13
Amazoniac nut
Combu, Assai, cocoa, 36 3.72 1.44 70.7 4.18 14.72
Jaguar, rubber
Islands,
Parad
Mishana, Timber, non 1 2.50 422.0
Perue timber
Tome aqu, Intensive 28 176.42 543.87 339.3 16.46 1.47
Para' Agroforestry
Palcazu, Timber 1 3,500.0
Peru"

Schwartzman, 1989
FUNTAC, UCLA, EDF, TEA, field project data
La Fleur, 1989
Anderson, and Jardim, 1989
Peters et al, 1989
Flohrschutz, 1989, cited by Anderson, 1992
Hartshorn, 1990

., cited by Hecht, 1992

The role of extractive activities is likely to change

in the future, especially because the long-term prospects of

wild rubber in the Brazilian economy are not secure

(Allegretti, 1990). The extractive reserves in Acre have

been sustained by a Brazilian tariff policy that has

effectively fixed the domestic rubber price at roughly three

times the average world price (Fearnside, 1989). However,

due to changes in the Brazilian policy of tariff protection,

domestic Brazilian rubber prices have fallen, and rubber

tappers are increasingly turning to farming to recover lost

income (Browder, 1992).

Perhaps the most important factor currently threatening

rubber production in the Amazon is the pressure of

successful plantations in non-Amazonian Brazil. By 1991,

about 180,000 hectares of rubber on plantations were

reported for Brazil (ITC, 1993), mainly in the states of

Bahia and Sdo Paulo. Browder (1992) estimated that

cultivated rubber supplies 60% of the Brazilian market.

These plantations, which have the comparative advantage of

being closer to the demand centers, will soon be able to

supply all of Brazil's domestic rubber needs at a reduced

cost, and the rubber extraction is sure to decline in a time

when free market policies call for subsidy removal

(Richards, 1993).

Brazil nut collection has generated an important part

of the income from extractive activities due to

international support for improving processing and

commercialization, particularly for improving conditions

among the rubber tappers near Xapuri. Due to the

seasonality of the harvests, Brazil nuts and rubber are

complementary activities with Brazil nut being collected in

the rainy season and rubber extracted in the dry season.

Brazil nut extraction gained in importance after the

rubber price crash in 1910. Because national consumption is

very low, it depends mainly on the international market.

According to the Sociedade para o Desenvolvimento Tecno-

ecol6gico (ECOTEC) (1993), Brazil nut today could reach a

world market share of $67 million due to the "substitution

effect"3 common among the nut markets. However, O'Donnell

(1990) noted that the Brazil nut market share has declined

to less than five per cent of the edible nut market, due

mainly to the growth in consumption of almonds and hazel

nuts. This fact could affect the market share of Brazil

nuts.

The international nut market is very dynamic,

especially as the "substitution effect" increases.

Therefore, besides the difficulties to increase or at least

maintain production to match demand, Brazil nut production

faces an old problem: the aviamento system. Advancing money

and/or commodities is a basic prerequisite for the aviamento

If for any reason there is a decline in the supply of a specific
nut, other nuts take up the slack.

system to succeed. According to Man-Producten Rotterdam

(1991), scarcity of money advances was responsible for the

serious crop reduction in 1991. In that year, Brazilian

shippers stopped advancing money for two reasons: first, as

local brokers found their debts accumulating, instead of

buying more Brazil nuts, they used the money advances to pay

those debts. As a result, nearly 50 to 60% of the contracts

were not delivered. Second, interest rates in 1991 were

very high (30% per month).

The main Brazil nut producers, Brazil and Bolivia, are

competing for market share. The United States imports about

50% of the Brazil nut traded in world markets. From 1991 to

1993, the USA has imported about 3,000 tons of shelled nuts

from Bolivia, as compared to less than 2,000 tons per year

from Brazil (Table 2.2). Partially, this has to do with the

fact that Bolivia's prices are lower than those of Brazil

(labor costs for shelling are lower in Bolivia). In the

case of in-shell nuts, the situation is quite different:

Bolivia has exported lesser quantities than Brazil, and

since 1993 it has exported only shelled Brazil nuts.

The future of Brazil nut extraction may also be under

threat due to a reduction of the resource base. Nepstad et

al., (1993), report a lack of regeneration and a scarcity of

juvenile trees in Acre. Possible causes include over-

harvesting of seeds, a reduced population of dispersal

Table 2.2. Shelled and in-shell Brazil nuts, by countries, imported by the USA between 1989 and 1993.

1989 1990 1991 1992 1993
Country Kilograms Value (US$ US $/Kg Kilograms Value(US$ US $/Kg Kilograms ValueU US U $/Kg Kilograms Value (US$) US $/Kg Kilograms Value (US$) US S/Kg
SHELLED BOLMIA 1,112,453 3,587,744 323 2,236.155 5.127.869 2.29 3,439,300 6,821.236 198 2.701,817 4,532,969 1.68 3.138,164 5,532,984 176
BRAZIL 1,907,232 4,577,015 2.40 1,668,769 4,288,013 2.57 524,908 1,184,227 226 2,032.565 3,938.828 1.94 1,972,267 4.237.139 2.15
PERU 658,424 2,006,964 3.05 702.157 1,564,362 2.23 251,828 634,015 2.52 134,928 273.203 202 431,993 1.014,695 235
OTHERS 341,667 1,138.936 3.33 422,070 984,456 2.33 376,369 788,591 2.10 631.437 1,261.376 2.00 533.579 1,214,257 2.28

IN-SHELL BOLIVIA 15.000 51,150 341 166,224 146,673 088 124,494 310,966 250 30.356 54,161 1.78
BRAZIL 1,643,198 2,111,946 1.29 6,252,275 5,809.302 093 4,461,252 4,014,957 090 5.607,973 4,791,040 085 4,028,499 4.290,861 1.07
PERU 267,335 794,855 2.97 100,503 259,166 2.58 130,560 303,441 232 108,294 212,574 1.96 59.637 142,226 2.38

OTHERS 17,100 56,496 3.30 62.781 40,144 0.64 32.065 66,581 208 17,173 29,653 1.73 0 0 0.00

Source: World Trade Press, 1994

agents, and low germination rates due to small felled-tree

gaps in Acre's forest.

Other researchers, though, believe in the success of

traditional rubber and Brazil nut production. In a study

comparing production value and ecological costs of different

land uses in Acre, Hecht (1992) found that on average

agriculture, livestock, and extraction generate an annual

gross return per hectare of US $1,200, US $1,350, and US

$1,320 respectively. However, production costs were not

considered in this study, and inclusion of these data would

probably change the results dramatically. Extractive

activities use more labor than agriculture or livestock

production, and their return per work-day is lower than that

of other more intensive practices (Table 2.1).

Since livestock raising has been considered as a little

productive but highly profitable activity, more than $1

billion of highly subsidized credit and fiscal incentives

have been applied to the sector (Browder, 1988).

Furthermore, tax exemptions, land speculation, and capital

gains associated with rising land prices have stimulated

this type of land use (Repetto, 1988).

The state of Acre is a good example of this. The years

from 1960 to 1980 saw major changes in land use and

distribution. Extractive activities declined both in area

(by 65%) and number of holdings, while ranching and

agriculture increased notably, mainly as a result of the

government's policy of frontier development through the

expansion of the road network, resource privatization, and

subsidies (Richards, 1993).

Starting in 1970, the Acre government supported an

intensive propaganda program in southern Brazil, mainly Sdo

Paulo, to attract both national and international capital.

Acrean lands were offered at very low cost, particularly to

livestock producers, and generous financing was provided

through SUDAM (Regional Development Superintendency) (IMAC,

1991).

"Land cleaning" operations, meaning the removal of

rubber tappers and small farmers, became common from 1973

onwards as ranchers intimidated rubber tappers and bought

their land at very low prices. This development, coupled

with the deterioration of rubber and Brazil nut prices, led

to an intensive outmigration from rural to urban areas.

Whereas in 1970, Acre had had 57,790 inhabitants in cities

(greater than 2,000 people), by 1980 the total population in

these cities had reached 131,000, and by 1991 the figure was

255,000. Rio Branco, Acre's capital, had seen an increase

from 35,000 people in 1970 to over 89,800 in 1991 and to

167,457 for 1991 (IBGE, 1991). The rural population

decreased from 56% in 1980 to 38% in 1991.

The urban migration process was accompanied by rapid

deforestation. The area deforested in Acre increased from

77,000 hectares in 1975 to 552,000 hectares in 1989 (INPE,

1989). In an attempt to improve the living conditions of

the rural populace, governmental and even some international

programs spent a lot of effort supporting agricultural

activities. However, the ecological and economic results

are far from viable.

Agricultural research in the Amazon has focused on

solving the constraints of agriculture using external

models, without considering more appropriate local

production models. Thus, most of the effort has been

directed at soil and environmental problems in an endeavor

to develop continuous cropping as the only alternative to

shifting agriculture.

In the 1960s an ambitious research program began in the

Peruvian Amazon with an agreement between the government and

the University of North Carolina, the goal being the

development of intensive agriculture in the Amazon based on

the knowledge that: (1) the Amazon has 345 million well

drained hectares, with slopes of a less than 30%, which

represent one of the world's biggest reserves for crops,

livestock, and forestry under rainfed conditions; and (2)

75% of the Amazon's soils are oxisols and ultisols, 15% are

entisols of fluvial origin, and only 6% of the well drained

soils (Altamira, Porto Velho, Rio Branco, and the eastern

Ecuador) have a high natural fertility Therefore, an

adequate knowledge of the soil's chemical properties was

imperative. The final justification for developing the

27

Yurimaguas project was the high environmental impact of the

agricultural system utilized by local farmers in forested

areas: their slash-and-burn method consisted of clearing the

forest and cropping the land until native fertility provided

by the ash was depleted, which took only a few seasons, and

then moving on to the next area (Cochrane and Sanchez,

1982).

Several cropping systems and time-of-planting studies

were conducted at Yurimaguas with maize, soybean, groundnut,

rice, and cowpea to determine the most promising systems of

three crops per year. Unfortunately, continuous monoculture

of the same crop did not produce sustained yields because of

the adverse effects of climate and a population increase of

various pathogens (Valverde and Bandy, 1981).

According to Bandy and Sanchez (1982), many of the

drawbacks of continuous monoculture were eliminated by using

crop rotation. The most promising results so far were

obtained by growing three crops per year on a rotational

basis (upland rice-maize-soybean or upland rice-

groundnut-soybean). After ten years, the average yields per

hectare were 2.71 tons of rice (37 harvests), 2.81 tons of

maize (17 harvests), 2.3 tons of soybean (24 harvests), and

3.46 tons of groundnut (10 harvests). Both production

levels and economic returns were acceptable.

However, because investment and the cost of inputs are

too high, these results are not applicable for most farmers

in the Amazon. The cost of the fertilization plan for one

year (Table 2.3) is US $875/ha (Cochrane and Sanchez, 1982).

These fertilizer levels do not differ substantially from

those required for maize, soybean, and groundnut cultivated

in the ultisols of the southeastern United States. The

problem with these figures in the Amazon, however, is that

fertilizers, lime, credit, and an adequate market system are

not available (Arguello Arias, 1988).

Table 2.3 Fertilizer requirements for continuous cultivation
of three crops per year in an Ultisol in Yurimaguas,
Peru.

Input Rate Frequency
Lime 3 tons/ha once every three years
N 60-100 Kg/ha rice and maize, respectively
P 25 Kg/ha every crop
K 100 Kg/ha every crop, split applied
Mg 25 Kg/ha every crop, unless lime is used
Cu 1 Kg/ha once a year or once every two years
Zn 1 Kg/ha once a year or once every two years
B 1 Kg/ha once a year
Mo 20 g/ha mixed with legume seeds
Cost of all inputs $875/ha/year'

a Cochrane and Sanchez, 1982
Source: Bandy and Sanchez, 1982

In spite of the enormous difficulties of implementing

the results of agricultural research in the Amazon, some

projects have attempted to encourage rubber tappers to use

this model as an alternative to extractive activities.

Projeto Esperanga (Project Hope) in the state of Amazonas

moved 100 local rubber tapper families to a planned 300

29

family settlement of agricultural colonists. After only two

years, 80 colonist families were left, due to the poor soil

conditions and the unsuitability of the agricultural

technology promoted by the project (Melone, 1988).

The deserted farms are usually bought by ranchers who

consolidate large tracts of land. Permanent employment on

these ranches is quite limited since livestock raising is

not a labor intensive activity. When the pastures become

degraded, the human carrying capacity of the area drops even

further. Most of the scientific community argues that most

of rural Amazonia simply cannot support a large population4

(Fearnside, 1989; Weischet and Caviedes, 1994).

At this point, it is clear that more intensive

agriculture could intensify difficulties for people who have

historically made their living from extractive activities.

Taking into consideration that the continuation of small

scale agricultural practices involves a lower risk for these

people and would keep the environmental impact at a minimum,

extractive activities might continue to be the main focal

point. This approach has attracted attention, especially

after the creation of extractive reserves in Brazil in 1990.

In spite of the lack of research on improving extractive

activities, some experiences show that there is economic

4There are however, some ecologically exceptional areas with
relatively fertile soils in the vicinity of Altamira, Porto Velho, part of
the state of Acre, and the Amazon floodplain.

potential for extractivism, particularly if developed in

areas where people have historically depended on it.

Alternatives for Improving Extractive Activities

According to O'Donnell (1990), there are four

alternatives for improving economic returns from extraction.

The first is to enrich the forest by planting economically

valuable plants and removing competitors, the second is to

utilize the natural high densities of certain tree species

which are scattered throughout the Amazon region, the third

is to increase the value added to forest products before

they leave the reserves, and the fourth is to increase the

number of species that are extracted.

Taking into account that extractive activities are not

isolated from subsistence agriculture and that there is

always some area of the production unit dedicated to this

activity, any improvement must cover both extractive and

agricultural activities. This realization leads to an

examination of the possibility for developing agroforestry

systems.

Another more recent alternative, pointed out by Clement

(1993), is the creation of income-generating forests in the

Amazon. This approach would involve most of the

alternatives considered above and could be applied both to

non-destructive harvesting of a large number of existing

31

species and to the introduction of species into the standing

forest or previously degraded fragments of it.

Enriching the Forest by Planting Economically Valuable
Plants and Removing Competitors

The experiences gained in the Peruvian Amazon

(Hartshorn, 1990) and Suriname (De Graff, 1982) could serve

as models for low intensity forest management. In these two

cases, the key to sustainable production in the different

types of tropical forest is clearly an adequate pace of the

forest's natural regeneration. There are two groups of

techniques that encourage such regeneration. The first

ranges from extreme treatments such as clearcutting, usually

in strips, followed by tending of natural regeneration, to

less drastic interventions that include the preservation of

a certain amount of cover ("shelter-wood") and, finally,

very conservative or cautious steps towards creating

openings within the forest (Budowski, 1988).

The second group of techniques combines natural

regeneration with artificial plantings that enrich the

forest with small amounts of valuable seedlings. Logging

practices often create favorable conditions for natural

regeneration, but at present most timber exploitation

schemes are not very concerned about existing seedlings,

saplings, and pole trees of desirable species (Budowski,

1988).

32

Hartshorn (1990) has been working with the first group

of techniques. His system produces systematic gaps in the

forest by clearcutting and carefully extracting the

resulting products in strips 20 to 50 meters (50 to 150

feet) wide. Only a small fraction of the forest is strip-

cut each year, allowing the gaps to close and making it

possible to return to the same plot after 30 years. The

system has been used in the Palcazd region of the Peruvian

Amazon, which receives abundant rainfall. Early results

show that there is abundant regeneration and initial growth

is excellent, not only of seedlings from adjacent trees, but

also in top regeneration of trees that were cut. Marketable

trees are locally processed in sawmills and treatment plants

for poles and posts.

This project is based on collective management by the

indigenous community, which owns the land. The long term

success of the activity will depend, among other things, on

keeping the community interested in all the management

operations, including logging, processing, and marketing,

which should be the case as long as the members receives

direct and continuous benefits from the managed resources

(Kishor and Constantino, 1993).

Utilizing the Natural High Densities of Certain Tree Species
Which Are Scattered Throughout the Amazon Region

The sustainable extraction of non-timber resources is

easier and significantly more profitable in high density

plant communities that include fewer species than in more

diverse communities (Peters, 1989). Anderson (1992),

Anderson and loris (1992), and Anderson and Jardim (1989)

report the results of research on Combu and Jaguar (Ilha das

Ongas) islands near Bel6m. The vegetation is floodplain

forest that is characterized by a high dominance of a few

tree species, many of which are of economic importance and

form the basis for extractive systems, the most notable

being the assai palm (Euterpe oleracea).

The harvest of assai fruits provides the residents of

both islands with their primary source of cash income, since

the Bel6m market is easily accessible. Other sources of

family income include lumber, heart-of-palm, shrimp, rubber

on Jaguar Island and semi-wild cacao on Combu.

The average size of a family holding on Combu is 36

hectares. A family generally uses less than one hectare for

a garden and to raise small livestock near the house. The

rest of the holding consists of managed forest, in which the

caboclos plant seedlings of valuable trees and eliminate

competitors through selective thinning. The costs of

production per hectare are approximately 18 labor days,

including one day of hired labor, and $1.44 in material

costs. The annual gross return is over $70 per hectare,

nearly $2,550 per holding, or $4.18 per work-day of family

labor. Again, this compares favorably with average wages in

Bel6m, which Anderson and Jardim (1989) estimated at US

$2.65 per day for 1986. Fewer than 20% of the Combu's

inhabitants are employed off the island (Anderson, 1992).

Anderson and his associates also tested the effects of

traditional forest management practices on the assai palm.

Fruit production from unmanaged forest provides returns of

$235 per hectare per year. Pruning the palms and thinning

their competitors requires approximately ten labor days per

year (an opportunity cost of US $26.50 for potential wages

in Bel6m) and results in a gross annual return of $370 per

hectare (Anderson and loris, 1992). This does not include

the return from the timber which is "thinned" or the heart-

of-palm hearts from the pruned stems.

Hoping to conduct a more systematic search for

extractive reserve sites, the Rubber Tappers Council has

begun a project with the Woods Hole Research Center to

analyze the distribution of forest resources and to identify

areas with high concentrations of economically valuable

species (O'Donnell, 1990). However, the mere existence of

forest resources does not guarantee the success of an

extractive reserve. If there is no functioning market for

forest products or no community organization in charge of

managing common usufruct rights, a reserve is unlikely to

succeed without significant outside assistance.

Increasing the Value Added to Forest Products Before They
Leave the Reserves

The Xapuri Agroextractivist Cooperative (CAEX), in

Acre, built a Brazil nut processing plant with support from

different organizations. In 1994, the CAEX paid US $0.64/kg

of nuts processed by the household units. This figure

applied to Brazil nuts that the CAEX had already purchased

from various gatherers during the harvest and had stored in

the central shed near a mini-factory. However, when

households that operated their own processing unit could

harvest and process Brazil nuts, the CAEX paid US $1.32/kg

(Campbell, 1996).

According to Hartshorn (1990), local processing of

timber for sawnwood, preserved posts and poles, and charcoal

adds considerable value to the native forest products.

Gross returns (after local processing) from natural forest

management by the Yanesha Forestry Cooperative in the

Peruvian Amazon are projected to be US $3500 per hectare

harvested. However, monetary assistance for infrastructure

and training in timber processing and management of these

projects is necessary if they are to succeed.

Increasing the Number of Species That Are Extracted

This initiative makes it necessary to create new

markets, or obtain access to existing ones, and to learn how

to identify, extract, process, and store the new products.

Increasing the Value Added to Forest Products Before They
Leave the Reserves

The Xapuri Agroextractivist Cooperative (CAEX), in

Acre, built a Brazil nut processing plant with support from

different organizations. In 1994, the CAEX paid US $0.64/kg

of nuts processed by the household units. This figure

applied to Brazil nuts that the CAEX had already purchased

from various gatherers during the harvest and had stored in

the central shed near a mini-factory. However, when

households that operated their own processing unit could

harvest and process Brazil nuts, the CAEX paid US $1.32/kg

(Campbell, 1996).

According to Hartshorn (1990), local processing of

timber for sawnwood, preserved posts and poles, and charcoal

adds considerable value to the native forest products.

Gross returns (after local processing) from natural forest

management by the Yanesha Forestry Cooperative in the

Peruvian Amazon are projected to be US $3500 per hectare

harvested. However, monetary assistance for infrastructure

and training in timber processing and management of these

projects is necessary if they are to succeed.

Increasing the Number of Species That Are Extracted

This initiative makes it necessary to create new

markets, or obtain access to existing ones, and to learn how

to identify, extract, process, and store the new products.

Non-timber products

There are some potentially valuable non-wood forest

products which are currently underutilized, such as wild

cacao, massaranduba (Manilkara elata), balata (Manilkara

bidentada), and piassaba. Sorvinha latex, which has

traditionally been harvested by destroying the tree, could

be sustainably tapped, leaving the tree to produce more

latex and fruits (Prance, 1989). Similarly, there are many

non-timber products that are successful on the international

market at present, or are likely to be so in the near

future. Examples are herbal antioxidants, which now

represent 10% of the antioxidant demand but are projected to

comprise the total demand for these products in the future;

natural food colorings, which now make up more than 95% of

the 100,000 tons in the food coloring market; medicines,

spices, and other natural products involved lately in green

consumerism (Duke 1992).

Oliveira et al. (1993) carried out a survey of non-

timber products used by nine communities of riberinhos in

the Tapaj6s National Forest. They inquired about the most

widely used non-timber species and products, their

collection processes, yield per tree, post-harvest

processing, and commercialization.

The researchers found that the riberinhos used products

from about 100 species; however, only 15 play an important

role in the community's economy. Among the latter are seeds

of cumaru (Dipterix odorata) and Brazil nuts; oils from

andiroba (Carapa guianensis) and copaiba (Copaifera

multijuga); bark of barbatimao (Stryphnodendron

pulcherrimum), pau d'arco roxo (Tabebuia impetiginosa),

preciosa (Aniba canelilla), and quinarana (Geissospermum

sericeum); latex of rubber, massaranduba (Manilkara elata),

amapd doce (Brosimum parinarioides), murure (Brosimopsis

acutifolia), and sucuuba (Hymanthus sucuba); and resin from

jatoba (Hymenaea courbaril).

Oliveira et al. (1993) made an estimate of the annual

production and the gross and net incomes for the riberinhos,

considering that they sell their products either at the

production site or in the Santardm market (Table 2.4). From

the selected 15 products, cumaru accounts for 50% of the

riberinhos' income, due to the high value of its seeds,

which are used on an industrial scale as an aromatizer for

tobacco, chocolate, and beverages and locally for medicines.

The figures obtained by Oliveira et al., based on

selling the 15 non-timber products at the production site,

show a potential net income of $435.16 per ha with a minimum

cost of $2.72 per ha. The figures are also based on the

following set of presuppositions: (a) every tree in the

studied area produces maximum yields; (b) the frequency of

the selected species is the same everywhere; (c) the species

are so closely spaced that one man could exploit 12 trees

per day; and (d) there is a complementarity in the phenology

38

of the species so that during the harvest season competition

for labor will not develop.

In spite of the promising income from the non-timber

species studied, Oliveira et al., (1993) underline that

these communities have their subsistence based mainly on the

practice of agriculture, hunting, and fishing. The apparent

lack of interest on the part of the communities to carry out

the extractive activities studied by Oliveira et al.

suggests that none or few of their presuppositions prevail

and that they overestimate the potential net returns of

these activities. Thus the results of Oliveira et al. have

to be considered carefully.

Medicinal plants

A study performed by Rider et al., (1994) in the city

of Cuiaba (state of Mato Grosso) in May and June of 1994

found a total of 162 vegetative species as sources for

medicinal products sold by 12 herbalists. However, 75% of

these herbalists had in common only 8.78% (15) of all

species found, and only 11 of the species found sold more

than 15 units (bags containing approximately 100 g) per

month, with the bark and seed of sucupira (Bowchisia

virginioides) being the most common (some 31.3 units per

medicinal herbalist per month). The study also indicated

that approximately 52% of the medicinal plants that make up

the marketed materials in Cuiaba are derived from wild

species found locally, while 13% come from Amazonia and 20%

Table 2.4 Non-timber products,
in the Tapaj6s National

and economic indicators, from "riberinhos" activities,
Forest, Santar6m, Para, Brazil.

Selling price Gross income Costs" (US $) Profit (US $)
(US $) (US $)
Common name Trees Product Unit Yield P S P S P S P S
/ha" /tree
Andiroba 4.6 Oil Liter 12 1.58 3.16 87.22 174.23 0.92 90.53 86.30 83.90
AmapA-doce 0.3 Latex Liter 15 0.79 1.89 3.56 8.50 0.06 3.83 3.50 4.67
Barbatimio 0.3 Bark Kilo 12 1.58 6.32 5.69 22.75 0.06 5.74 5.63 17.01
Copaibac 0.2 Oil Liter 80 1.58 3.16 25.28 50.56 0.04 26.24 25.24 24.32
Cumari 0.2 Seed Kilo 350 3.16 11.84 221.20 828.80 0.04 222.11 221.16 606.69
Brazil nut 0.7 Nut Liter 125 0.47 1.58 41.12 138.25 0.14 46.37 40.98 91.88
Jatobi 0.6 Resin Kilo 30 0.79 2.37 14.22 42.66 0.12 14.45 14.10 28.21
Massaranduba 2.2 Latex Kilo I 0.79 1.58 1.74 3.48 0.44 1.77 1.30 1.71
Mururd 0.9 Latex Liter 6 0.79 2.37 4.27 12.80 0.18 4.59 4.09 8.21
Pau-d'arco-roxo 0.3 Bark Kilo 35 0.79 3.16 8.30 33.18 0.06 8.44 8.24 24.74
Piquia 0.2 Oil Liter 10 1.58 4.74 3.16 9.48 0.04 3.28 3.12 6.20
Preciosa 0.2 Bark Kilo 10 0.16 1.18 0.32 2.36 0.04 0.34 0.28 2.02
Quinarana 1.3 Bark Kilo 10 0.79 3.16 10.27 41.08 0.26 10.44 10.01 30.64
Rubber 0.6 Latex Kilo 3 1.58 3.16 2.84 5.69 0.12 2.86 2.72 2.83
Sucuuba 1.0 Latex Liter 11 0.79 2.37 8.69 26.07 0.20 9.25 8.94 16.72
Total 437.881 ,400.04 2.72 450.34 435.16 949.75

P = at the producer site;
a DBH > 30 cm

S = at Santar6m

b Producer cost = labor cost
Santar6m cost = selling price plus transport cost
c Production per tree at the first extraction. In a second extraction (after six months)
it is 50% reduced
Source: Oliveira, Tavares, and Ferreira (1993)

are cultivated species. The average herbalist sells

approximately $500 worth of medicine per year, the clients

being predominantly poor, elderly, and female.

Timber products

It is well known that timber is one of the most

valuable products from the Amazon. However, timber

harvesting presents difficulties due to a lack of planning

in the efforts to keep both the ecosystem functioning and

the communities prospering.

The high value of timber and other forest products

should help ease the transition toward sustained yields, but

the shift will require drastic changes in official policies

for timber concessions, and it may be necessary to offer

incentives to those groups that take effective steps in that

direction (Budowski, 1988). Therefore, the challenge is to

find adequate ways to do it.

In smaller countries, where tropical forests have
been all but destroyed by loggers and follow-on
farmers, the cessation of tropical timber
extraction might be justifiable. However, the
omission of timber resources from discussions of
sustainable extraction in the case of Brazil seems
incomprehensible. Browder, 1992 :181.

Although there is little scientific literature about

successful experiences with systems based on sustainable

timber extraction in the Amazon, there is a potential for

this if properly managed.

Forest management and traditional methods of

exploitation. Many of the small timber producers in the

state of Amazonas use traditional methods of timber

extraction which are responsible for about 80% of the

state's timber production. While these methods have some

limitations, such as low production capacity and dependence

on family labor, from a social point of view, they have

guaranteed subsistence and land tenure for thousands of

families over generations.

The main characteristic of the traditional methods is

their low level of investment and technology. Although

productivity and economic gains are low, the environmental

impact, compared with mechanical exploitation, is minimal

and does not compromise floristic composition and natural

regeneration (Oliveira, 1992). Traditional timber

extraction uses separate methods for flooded and non-flooded

areas.

The harvest of commercially valuable logs in flooded

forests takes advantage of the river floods for extraction

and transport, thereby foregoing the use of heavy machinery.

This system depends greatly on the extraction and commercial

use of low-density logs and does not offer large volumes of

logs with high densities.

All of the techniques employed are based on

generations-old experiences adapted to current situations

and the technology available. Extraction is regulated by a

calendar that allows for the small producers' other

essential activities (hunting, fishing, extraction of non-

wood products, and subsistence agriculture), thus allowing

for a multiple use system of natural resources. Capital

investment is low, reduced to chainsaws, gasoline, a boat

(canoe), and food.

Producers are not organized in labor unions or

cooperative associations. The majority work independently

and alone. Also, there is no structured price control for

producers.

The traditional method of havesting timber in the

flooded areas has four phases: 1) opening trails and marking

trees; 2) cutting down trees; 3) transportation and

"cubicaje" (packaging); and 4) marketing.

Trails are opened with the help of machetes and without

instruments for orientation. Trails are determined by the

possibilities for access to and transport of logs during the

flood stage. Producers thus choose a main entrance to

streams, creeks and lakes that allows access and transport.

The cutting of trees begins after the trails are

opened. The intensity of the tree cutting process is low,

mainly because the selection is restricted to a few species

and large trees.

The transport phase is the most difficult. The log may

be guided with the aid of pushers (sticks forked at one

end), canoes, or paddles, depending on the water depth.

Once the logs are by the river banks, buyers measure

the logs, subtracting five cm from each log's diameter to

43
account for the thickness of the bark. Normally, only logs

with a diameter greater than 35 cm (without the bark) are

taken into account for packing. In July 1989, prices per

cubic meter were about $8 for logs brought to the river

banks and $25 in the sawmills.

As is the case in rubber extraction, the process is

financed by the buyers (patrons), who advance payment in the

form of shelter, oil, gasoline, and medicines at the

beginning of the extracting season. All of these goods are

supplied at high prices, coupled with the patrons' exclusive

rights for selling the product (at low prices).

The traditional method adapted to non-flooded forests

is similar, but there is a difference in the mode of

transportation. Since there are no waterways that could be

used for the transport operations, it becomes necessary to

process the logs at the original site. There, chainsaws

transform the logs into sawnwood. In special cases, a

producer may have available some animals to transport the

logs to sites with more sophisticated processing equipment,

such as a portable saw, which considerably increases

processing efficiency.

Environmental impact. Even though this type of logging

system has a "bottleneck" caused by unequal production

arrangements (similar to the aviamento), it may be

ecologically sustainable. Oliveira (1992) gives examples of

this system of logging production in the state of Amazonas,

where he analyzed the process of floristic succession in

gaps incurred by the exploitation and studied the structure

and future possibilities of natural regeneration and

sustainable production.

The study was performed on a 1,200 hectare property

exploited for more than 20 years in Abufari County

(municipio) in the state of Amazonas. The size of harvested

areas varied from 15 to 50 hectares. The planned cutting

cycle comprised 25 years, during which an average of 20

trees were cut per year (Table 2.5).

Table 2.5 Harvest intensity per year and area following
traditional exploration methods in Abufari, Amazonas,
Brazil.

Year Harvested Total of Harvested
area (Ha) harvested trees/Ha
trees
1977 48 17 0.4
1982 33 23 0.7
1986 15 16 1.1

Source: Oliveira, 1992

The author observed that the gaps always included a

great majority of those species that form part of the

primary forest. During the first year after cutting, there

was a sizeable reduction in the natural regeneration

population of the 1.5 m height-5.0 cm DBH classes. The

remaining classes were affected less. Regeneration depended

basically on the remaining classes in the 1.0-1.5 m height

category, on the classes begun from seeds stored in the

soil, and on those available near the gaps.

Perturbances promoted by harvest were limited to the

areas cleared in locating the trees and to the gaps formed

by fallen trees. There was an adequate stock of species

involved in the harvest process and, with the exception of

Ocotea cymbarum, all underwent sufficient natural

regeneration (Table 2.6). An average of about seven

commercial trees remained per hectare, a good indication

that the forest under observation was responding adequately

to the management practices.

Table 2.6 Natural regeneration after 10 years, following
traditional methods for timber exploration in Abufari,
Amazonas, Brazil.

Species DBH
25- 45- 65- 85- >100
45 65 85 100
Callophyllum 1.42 -
brasiliensis
Aniba hostimaniana 0.47 -
Nectandra sp. 0.83 0.56 -
Ocotea cymbarum -
Manilkara sp. 0.55 0.47 0.95
Virola sp. 2.37 -
Total 5.09 1.1 0.47 0.95

Source: Oliveira, 1992

Policies for forest management. Forest management can

also be applied to mechanized commercial harvesting.

However, costs impose a serious limit. Barreto and Uhl

(1993) calculated the cost for a cutting cycle of 20 years

in Paragominas (Par& State) at US $187 per hectare. To

encourage forest management, the harvest tax should be close

to this value. However, in the case of Paragominas, every

hectare harvested but not reforested is only taxed $55, so a

policy of forest management adoption is discouraged.

In Costa Rica, Kishor and Constantino (1993) found

that, although sustainable natural forest management is a

declared national goal, it is seldom put into practice.

Instead, private landowners prefer to mine the forest or

convert it into cattle ranches or forest plantations, since

natural forest management cannot compete financially with

alternative types of land use.

To make natural forest management an attractive option

for private landowners, Kishor and Constantino (1993)

propose high rates of real price appreciation and/or

subsidies, but, based on past trends and the availability of

substitutes, this is unlikely to happen.

They also found that consumer willingness to pay for

"green-labeling" is insufficient to cover the cost of

sustainable management. According to these authors, if

environmentally sustainable forest management is chosen as

the preferred land use from a social point of view,

subsidies are necessary. However, they conclude that, from

a budgetary efficiency point of view, in a situation of high

discount rates due to macroeconomic risks, it would make

more sense to target the subsidies to small farmers, who

face lower opportunity costs from doing sustainable

management than big or corporate owners.

Besides forest management, the efficiency of processing

needs to be increased. On average, the efficiency in

traditional lumber production is about 40%, and in improved

sawmills it is about 60%, which evidences much waste in the

two production systems. An alternative way to use these by-

products would be to develop small products for specialty

markets. In 1993, the USA imported $14,686,453 in small

wooden articles from Brazil (Table 2.7).

The production and marketing of tropical timber.

Tropical wood production represents only 14% to 15% of the

total timber produced globally. Production growth rates

during the 1980s fell to 1.5 percent, in comparison with 5.6

percent between 1960 and 1980. Table 2.8 shows a sizeable

increase in industrial roundwood (logs and pulp) and

processed wood products in most regions during 1960s and

1970s, followed by a considerable decrease in production

growth during the 1980s. An exception has been the

production of wood-based panels, in particular plywood,

manufactured in Asia (Panayotis et al., 1993).

The market for tropical timber is often seen as being

responsible for the disappearance of tropical forests. But

these charges are not justified on two accounts. First,

deforestation of the tropics is mainly caused by agriculture

and cattle raising. According to FAO (1987), 60.4% of

Table 2.7. Brazilian manufactured wood-based articles imported by the USA between 1989 and 1993.

ARTICLE 1990 1990 1990 1990 1990
Units Value (US$) US$/unit Units Value (US$) US$lunit Units Value (US US$/unit Units Value (US$) US$/unit Units Value (US$) US$lunit
CIGAR AND CIGARETTE BOXES, OF WOOD 2,000 6,760 3.38 12.600 46.315 368 11.400 28.219 2,48 3,881 11.514 2.97
JEWELRY BOXES. ETC WOOD 48,471 97.575 2.01 137.500 210,480 1.53 61,681 94.826 1.54 46.438 193.195 4.16 175.390 551.627 3.15
TABLEWARE AND KITCHENWARE. OF WOOD 44.576 18.685 111.412 265.070 992.123
FORKS AND SPOONS, OF WOOD 75.119 21,344 108.330 825.258 153.793
STATUETTES AND OTHER ORNAMENTS. OF WOOD 6.054 4,500 20.115 33.457 199.651
WOODEN FRAMES PAINTINGS. PHOTOGRAPHS 360.471 369,966 195.562 127.125 107.205 89.170 083
ARTICLES OF WOOD 9.849.068
GUITARS NOT OVER US $100 2.360 59.760 25.32 1.100 38.850 35.32 738 27.666 37 49 4.390 179.012 40.78
GUITARS 14 3.012 215.14 14 6.548 467.71
WOODEN FURNITURE OF USED IN THE KITCHEN. 69.801 238.407 21.956 207.813 86.361
WOOD MARQUETRY, INLAID WOOD. CASKETS, ETC 63.489 308.294 157.538 145.452 649,224
WOODEN FURNITURE USED IN OFFICES 933.925 695.325 633.002 742,734 970.196
TOOTHPICKS OF WOOD 420.238 366,127 372.324 411.849 379,155
SKEWERS/CANDY STICKS & SIMILAR 139,490 206.310 252,193 242,553 257.194
PICKETS. PANGS, POSTS AND RAILS. SAWN 6,804 311,817

TOTAL 52.845 2.280.270 138.600 2.478.288 74,281 2,020,377 58,576 3.250,391 290.880 14.686,453

Source: World Trade Press, 1994

global deforestation is caused by agricultural practices

performed for subsistence, 16.8% by permanent agricultural

practices, 8.3% by cattle raising, 5.6% by lumbering

activities, and 7.9% by the production of firewood and

charcoal. Second, timber exports, particularly of roundwood

and sawnwood, constitute only 11% of the total timber

production. In the Brazilian Amazon, from 25 million ha of

land deforested during the '70s and '80s, at least 70% were

for planting pasture for cattle raising, while the rest were

deforested as consequence of shifting agriculture and

extractive activities (mainly logging), in that order of

importance (Mahar, 1989).

Table 2.8 Timber products exported
and 1990.

However, this does not mean to

is not a serious problem. In fact,

by region between 1961

say that deforestation

a regional analysis

1961 1970 1980 1990
Product/Region In 1,000 cubic meters
Industrial Roundwood
Asia 21,606 47,690 101,617 118,45
Latin America 20,632 28,457 54,735 63,771
Africa 18,632 26,989 34,493 38,604

Sawnwood
Asia 4,295 7,092 27,004 37,504
Latin America 6,273 8,621 13,708 17,063
Africa 1,816 2,612 5,169 5,780

Wood-based Panels
Asia 104 1,079 5,678 14,461
Latin America 648 2,056 4,221 5,034
Africa 155 523 1,096 1,422
source: FAO, 1991

would show that, in reality, the problem of deforestation

has had a negative influence on timber production in the

tropics, with regional differences of course. Table 2.8

shows a decline in the growth rate of tropical timber

production during the 1980s. This decline was caused

fundamentally by the depletion of African and Asian forests.

Africa, with one of the world's highest deforestation

estimates, lost 11% of its market share. In Asia, the

Philippines, which during the 1960s was among the principal

timber exporters, today imports timber (Panayotis et al.,

1993).

In this context, and according to statistics provided

by ITTO (1993), Latin America emerged during the last decade

with the highest increase in exports of tropical products:

from 614 million cubic meters in 1978 to 2,633 million in

1991, this was a spectacular growth of 429% in 13 years.

In spite of the fact that the Asian countries

(particularly Malaysia and Indonesia) continue to be the

main exporters of timber (80% of global exports), Asia is

also the most important importer of timber traded on world

markets, with more than 54% of the total market, the main

countries being China, Japan, and South Korea (Panayotis et

al., 1993).

The depletion of forest resources in Africa and Asia

and restrictive policies concerning the exploitation of

roundwood or sawnwood timber by some of the most important

exporters make the tropical rain forests of South America

the main areas for future production of timber.

Even though tropical America, especially Brazil,

competes in the market with many types of timber products,

as seen in Table 2.9, it has been concentrating on sawnwood

and plywood.

Table 2.9 Destination of timber products exported from Bel6m
(Par&), Brazil, in 1993.

Sawnwood Plywood Veneers Panels Total %
USA 59,766 92,778 18,207 1,791 172,54 25.92
North Europe 84,568 7,479 828 897 93,772 14.09
The Caribbean 33,174 54,157 3 3,391 90,725 13.63
England 50,161 20,467 56 488 71,172 10.69
Persian Gulf 14,531 30,789 281 1,066 46,672 7.01
Philippines 37,816 0 0 0 37,816 5.68
Mediterranean 28,487 823 469 3,793 33,572 5.04
Portugal 18,925 38 127 6,444 25,534 3.84
Japan 3,134 11,029 1,170 0 15,333 2.30
Mexico 13,397 255 0 176 13,828 2.08
Ireland 8,654 1,316 0 160 10,130 1.52
Canada 4,164 0 31 18 4,213 0.64
Others 38 9,606 1,207 1,051 50,354 7.56
Total 395,272 228,73 22,379 19,275 665,66 100.0
Source: AIMEX, 1994

Aqroforestry

Montelo Moura (1994) studied the feasibility of

agroforestry systems with species of economic potential in

extractive reserves. Using linear programming, he

determined the effects on gross income of production systems

characterized by mixed subsistence crops and plantations of

rubber and Brazil nut trees (Table 2.10). He ran tests on

four scenarios: (1) the current situation, defined as the

current productive system with its traditional extractive

Table 2.10. Gross returns and activities in the production optimum plans for the current,
and alternative situations in the Cachoeira extractive settlement, Acre,Brazil

Situation GR ERB EBN SA SA2 RRB RBN CMD01 CMD02 CMD03 CMD07 CMD09 CMD10 CMD011 CCC CCI
US $ ha ha ha ha ha ha work-day work-day work-day work-day work-day work-day work-day US $ US $
CS 2,087 251 330 1
SI 2,517 330 330 1.7 7.5
SII 2,886 330 330 3 24.5 65 32
SIII 7,829 330 330 3 3.5 16.5 401 401 521 157 21 111 90 4,501 1,796
Sllla 7,473 330 330 3 3.5 12.5 308 308 425 117 21 111 90 3,653 1,375
Slll.b 7,120 330 330 3 3.5 8.59 216 216 329 78 21 111 90 2,812 956
SIllc 6,768 330 330 3 3.5 4.67 125 125 233 39 21 111 90 1,974 539
Silld 6,445 330 330 3 3.5 0.75 34 34 137 21 111 90 1,135 121

CS= Current situation; SI=Situation I ; SII = Situation II, etc.
GR= Gross return
ERB= Rubber extraction
EBN=Brazil Nut extraction
SA=Subsistence agriculture; SA2= Subsistence agriculture recommended
RR=Rubber reforestation
BNR=Brazil Nut reforestation
CMD01= Hired labor month 01; CMD02= Hired labor month 02; etc.
CCC=Initial capital costs
CCI=Capital investments

Source: Montelo Moura, 1994.

practices and resource availability; (2) Alternative I,

similar to the current situation, but hiring labor during

the most labor intensive months; (3) Alternative II, similar

to Alternative I, but with the option to use both

recommended technologies and credits to cover costs and

investments; and (4) Alternative III, similar to Alternative

II, but introducing plantations of rubber and Brazil nut

trees. Alternative III was further tested at four credit

levels: at 80%, 60%, 40%, and 20% of the credit needed.

After running the linear programming simulation,

Montelo Moura (1994) found that Alternative III at the 20%

credit level would be the best alternative, where a

production unit using $1,256 would achieve a gross income of

$6,415, using 1,643% more labor than in Alternative II.

Additionally, this was less risky than the other

alternatives using credit and showed the best internal rate

of return (115%). However, the problem with this

alternative is the procurement of the hired labor and the

credit. It is also not clear how the author envisions

financing the rubber and Brazil nut plantations. It is

known that the cumulative variable costs for Brazilian

plantation rubber are about $1,854 per hectare for the first

seven years (the immature period) (ITC, 1993); the figure

for Brazil nuts could be higher, since the period before

production begins is longer.

54
A search in the international market for products that

occur naturally or could be cultivated in the Amazon can

help identify potential species to be incorporated into

agroforestry systems. After Brazil nut, pepper (black and

white) is probably the second most important non-timber

forest product exported to the USA from the Amazon,

constituting some 30% of the $20 million total Brazilian

exports of this product between 1989 and 1993. However,

prices have been dropping recently, and this is also true of

products such as annatto and refined babassd oil. In

contrast, other products, such as herbs used for medicinal

or other purposes, have risen in quantity marketed and

maintained their prices, while others, such as fruits of the

genus Capsicum, have even increased in price (Table 2.11).

The potential international and national market for

products from selected agroforestry species for the state of

Acre has been researched by ECOTEC. The products were

selected based on the recommendations of various

institutions engaged in agroforestry research systems in the

state of Acre. They involved food products such as assai

(Euterpe sp.), cupuassG (Theobroma grandiflorum), guaran&

(Pranceara cupana), heart-of-palm (from Euterpe oleracea),

peach palm (Bactris gasipaes), and annatto (Bixa orellana);

nuts such as Brazil nut and cashew (Anacardium sp.);

commodities like rubber, coffee, cocoa, and black pepper

Table 2.11. Brazilian natural or cultivated non-timber products imported by the USA, between 1989 and 1993.

190 I 101 I IO^e I l'JOJ

Kilograms Value (US$ USS/kg Kilorams Value (US$ USS/kg Klograms_ Value US U ale US$) USk Kilograms Value (US) US$/kg Kilograms Value (US$ USS/lk
ANNATO, ARCHIL, COCHINEAL, CUDBEAR, LITM 58,000 133,900 2.31 17.391 13.913 080
PALM KERNEL OR BABASSU OIL, REFINED 1,080 2.084 1.93 1,074 2.915 2.71 6.660 12.559 189
PALM KERNEL OR BABASSU OIL, CRUDE 162.271 82,941 0.51 47,440 77.872 1.64 28,800 34.442 1.20 0 10,800 19,488 1.80
NUT OILS NESOI VWETHER/NOT REFINED 128,922 64,000 0.50 42 1,300 30.95 304.208 215.938 0.71 200.100 223.876 1 12
PEPPER (GENUS PIPER) NOT CRUSHED OR GROUNDED 11,037.754 21,417.585 1.94 8,777.468 13.932.339 1.59 14.907,703 18,268.916 1.23 6,601,015 6,865,735 1.04 4.579.556 5.531.682 1.21
PEPPER (GENUS PIPER) CRUSHED OR GROUNDED 37.000 180.400 4 88 16,500 36.638 222 1,500 3.492 2.33 51.383 84.263 1.64 322.000 817,942 2.54
PLANTS & PARTS EX MINT LEAVES USED AS HERBS 33.0981 60656 1.83 19,051 29.181 1.53 18,543 29,110 157 67.103 158.271 2.36 101.056 197.835 196
PLANTS & PARTS, USED PRIMRLY IN PHARMACY 59,223 130.940 2.21 88.497 215.673 2.44 43.384 141,056 325 52.271 167,17 3.21 144.260 583,982 405
FRUITS OF THE GENUS CAPSICUM, NOT GROUNDED ___ 1001 3,252 32.52 15,150 28.500 1.88 46.179 144.540 3.13 20.097 143.430 7.14

Source: World Trade Press, 1994

I

1990 1991 1 IUYZ I IuyJ

I

(Piper nigrum); oils such as copaiba and dendd (Elaeis

guineensis); and lumber.

Based on secondary information, ECOTEC (1993) evaluated

the market share, price flexibility, price stability, the

price oscillation index, and the risk coefficient of each

product. In order to establish the price stability

coefficient, the market share of each product is divided by

the income per work-day, the result of which is divided by

the price flexibility index (the inverse of price elasticity

of demand). At the same time, market potential was

evaluated by considering the change in supply according to

price at the international, national, and local levels. A

risk coefficient was also determined by dividing the income

per work-day for each product by the price flexibility index

and dividing this result by the price oscillation index

(yearly average of price oscillation). In other words, a

low risk will be the consequence of high wages, a high

flexibility index, and a low price oscillation index.

Based on its findings (Table 2.12), ECOTEC (1993)

divided the studied products into three groups: (1)

recommended products: (Brazil nut and rubber); (2) products

in need of complementary analyses assaii, cashew, cupuassu,

heart-of-palm, and lumber); and (3) products not recommended

(the commodities, copaiba oil, guarand, peach palm, and

annatto).

Table 2.12. Risk analysis performed by ECOTEC on 15 potential cash crops for the Amazon

World market Brazilian market income ($ per work-day) Price/supply Price stability
Item In million US$ In million US$ extraction cultivated flexibility coefficient risk

Rubber 4,100.00 52.00 2.74/300 ha 3.9 low optimum very low
Brazil Nut 33.00 27.00 3.3/300 ha 6.88 very low medium low
Copaiba 1.00 1.00 2.81 high very low highest
Assai 32.40 2.65 medium medium medium
Heart-of-palm 8.30 1.13 medium very low medium
Cupuassu 1.50 2.35 high very low very high
Annato 9.50 5.60 2.75 high very low high
Guarana 11.20 9.70 1.13 medium very low high
Black pepper 223.30 35.90 3.36 high medium high
Peach-palm 1.30 1.56 very high very low high
Cacao 2,266.00 323.80 3.49 high good medium
Dende 3,201.00 7.50 2.39 low optimun lowest
Coffee 11,194.00 3,629.00 3.22 high good low
Caju 378.00 114.30 2.78 very low good low
Sawn timber 13,209.00 160.50 0.91 high optimun very low

Source: ECOTEC, 1993

There are, however, two observations concerning the

ECOTEC findings. The first has to do with the way it

compares the income per work-day of the selected products.

For rubber and Brazil nuts, the value referred to income

generated from 300 hectares, while for the rest of the

products, the value referred to that generated on only one

hectare. Thus, the risk coefficient for extracted rubber

and Brazil nuts was -2.74 and -4.70, respectively, but when

income per work-day was applied to one hectare, as in the

case of the other products, the risk value was only -0.009

and -0.011 for rubber and Brazil nut, respectively.

Consequently, the initial finding of 'low risk' has to be

reversed to 'the highest risk of all studied products.' In

other words, rubber and Brazil nut are not good bets.

The second observation is related to the productivity

figures and lumber prices. ECOTEC considered a productivity

of one cubic meter of lumber per hectare of cultivated

forest and a market price of $9.05 per cubic meter.

According to other sources, these figures are too low,

making the ECOTEC results appear biased.

Creating Income-Generating Forests in the Amazon

Also of interest for conservation in the Amazon is the

creation of income-generating forests. Myers (1984)

proposed an "industrial forest" to conserve biodiversity in

the tropical biome while obtaining economic returns from the

59

forest. Clement (1993) prefers the term 'income-generating'

to 'industrial,' and he also emphasizes that this

alternative can be applied to both non-destructive

harvesting of large numbers of existing species and the

introduction of species with new uses into the standing

forest or previously degraded fragments of it. However, an

income-generating forest differs from an agroindustrial

plantation in that biodiversity is permitted or even

encouraged, while in a plantation only the economically

exploitable species are permitted.

Clement and Myers agree in the sense that only those

species with established markets or with markets that could

be developed by entrepreneurs would be harvested or planted,

but the forest could contain numerous other species with

potential for use at some point in the future. However, to

Clement, the question is how to design and create an income-

generating forest that would enhance the economic viability

and management of its sustainable use? He recommends: a)

deciding on the number of economically exploitable species

to be included while trying to expand the few species

currently used; b) the methodological selection of species

from a minimum of 100 with known uses and several thousand

with potential uses, taking into account both market

orientation and species adaptation to the local environment;

and c) the design of procedures for species introduction

into and/or management in the forest ecosystem.

60

Today, most extractive reserves are restricted to only

two products (rubber and Brazil nut), which is one of the

reasons for their economic fragility. Colonists and

caboclos tend to depend upon one or two 'export' crops and a

few subsistence crops. The Amerindians rarely cultivate

'export' crops, although they may be involved in rubber and

Brazil nut extraction to some extent (Clement, 1993).

Diversification of production is desirable so that, if the

market for one of several products declines drastically, the

producer still has other products on which to rely.

Another strong reason in favor of diversified forest

production is that any extractive species that develops

attractive economic prospects will immediately be put into

cultivation elsewhere and produced more cheaply there (Homma

1989). The history of modern tropical agriculture clearly

corroborates this observation. Rubber extraction was

restricted to the Amazon until its value rose to a level

that made its cultivation of interest in Southeast Asia

(Hecht and Cockburn, 1990). Quinine (Cinchona ledgeriana)

followed the same route in the late 1800s (Dominguez and

Gomez, 1990). The only way out of this dilemma consists of

finding and developing large numbers of species in a process

that continues over long periods of time. The long term

nature of this process has probably been underestimated by

those starting to work on the development of new species for

extractive purposes (Clement, 1993).

Diversification of species also has its limitations.

According to Clement (1993), these are: 1) knowledge about

the species; 2) markets; 3) labor force; and 4) capital.

Generally speaking, at the producer level, only a few

species can be successfully managed by each family. At the

community level, though, it should be possible to increase

this number.

Clement (1993) emphasizes that the selection of new

species will have to be highly site specific. He also

develops several criteria that have to be met by the new

products, which should be: (1) of high unit value, (2)

easily processed, and (3) easily stored, either before or

after processing.

Arkcoll and Clement (1989) suggested that many

Amazonian fruit crops might also have potential as new

crops. These can only be stored, however, if they are first

processed or pre-processed (cleaned and frozen, for

example). Most fruit crops meet the first criterion, but

the second and third ones can only be met after the

infrastructure for processing and storage has become

available. In other words, they will be suitable for most

areas in the Amazon only at some point in the future, when

practitioners of the income-generating forest have become

successful.

Finally, Clement (1993) discusses how to establish the

income-generating forest, taking into account whether an

area has been submitted to much, little, or no

deforestation. This is a very important aspect, especially

in view of the fact that, in the Amazon, each household has

developed its own way of taking advantage of the different

deforestation patterns.

Strategies for Enhancing Income Generation

Experiences with communities concerning strategies for

enhancing income generation from the Amazon are rare.

Perhaps the most important is the experience of Cultural

Survival Enterprises (CSE) which demonstrates that forests,

in the long run, have more value left standing than razed

for pasture or conventional agriculture. Scientific

research and subsequent economic projections suggest that

forest areas are ultimately worth more for their sustainably

harvested timber and non-timber products than when cleared

for logging and/or conversion to pasture, agriculture, or

even plantations (Clay, 1993). Since 1990, CSE has sent

about 350 different non-timber forest products (NTFPs) out

of the Brazilian Amazon for sampling by 120 companies in the

USA and Europe.

To provide a theoretical and practical strategy for

improving the marketing of NTFPs, there are eleven major

lessons CSE has learned during this period: (1) define land

and resource rights, (2) start with products already on the

market, (3) increase competitiveness, (4) diversify

production and reduce dependence on a few products, (5)

63

diversify the markets for raw and processed forest products,

(6) add value locally, (7) capture the value added to the

products as they travel farther from the forest, (8)

commercialize high volumes, (9) aim for a decent profit, not

a killing, (10) solutions must be on the same scale as the

problems, and (11) international markets are meant to

protect ecosystems, not the people who live in them (Clay,

1993).

Conclusions

The importance of extractivism based on rubber and

Brazil nut collection has tended to decrease in the Amazon.

However, extractive activities still are practiced by an

important number of the Amazon's inhabitants. These

activities are performed based on unequal, although not

necessary exploitative, production arrangements, due to the

costly and risky transport conditions and the high risks of

marketing extractive products that normally have large price

variations.

Forest extraction based only on rubber and Brazil nut

is an inefficient land use practice. This inefficiency is a

consequence of: (1) the relatively large holdings that

families must exploit, (2) the relatively low productivity

of each production unit (for example, when compared with

rubber plantations), and (3) the relatively low income

64

generated by these activities when compared with other, more

intensive, uses of resources in the Amazon.

There are, however, some alternatives for improving

extractive activities in the Amazon, which include: (1)

enriching the forest by planting economically valuable

plants and removing competitors, (2) utilizing the natural

high densities of certain tree species which are scattered

throughout the Amazon, (3) increasing the value added to

forest products before they leave the reserves, (4)

increasing the number of species which are extracted, and

(5) developing agroforestry systems. Most of these

alternatives could be part of a much more recent approach

called the income-generating forest.

Both non-timber and timber products may be harvested in

the Amazon under sustainable management. There are

experiences which show that non-timber products (such as

Brazil nut, heart-of-palm, cumaru seeds, and essential oils)

and timber products (particularly lumber), when harvested

sustainably, may generate enough income to improve the

livelihood of people living in the Amazon forest.

One of the main strategies of developing an income-

generating forest is the diversification of the productive

base. Having the possibility of marketing several products

is less risky than marketing only one. However,

diversification also has limitations, mainly related to the

lack of experience, market, labor, and capital; in

65

consequence, only a few products may be successfully managed

at the family level. Selection of these products has to be

highly site specific, and the species selected must meet

conditions such as: (1) yield high unit value products, (2)

yield an easily processable product, and (3) yield an easily

stored product.

The dilemma summarized above shows that, in spite of

the difficulties of extractive activities, they still

constitute an important land use, particularly in forest

areas where reserves have been established in the Amazon.

Since conservation is one of the major objectives of these

reserves, and land use systems involving deforestation are

not allowed in these areas, improvement of extractivism

could help reconcile the objectives of conservation and

livelihood improvement of the people living there. To

achieve this, it is necessary to perform site specific

evaluations of alternatives to improve extractive activities

in the Amazon, which constitutes the main goal of this

research.

CHAPTER 3
STUDY SITE AND METHODOLOGY

Location of Study Site

The Antimari State Forest (ASF), located 80 km north of

the state's capital Rio Branco, was part of the former

Arapixi, Pacatuba, and Mapinguari rubber estates. The

forest extends over an area of 66,168 ha (FUNTAC, 1991a)

between the meridians 68000'19" and 68021'45" W, and the

parallels 09010' and 09031'S (Figure 3.1). In the flat to

lightly undulating terrain lie the settlements of 48

families (300 people), who live and work there, and another

25 families (150 people) within the area of influence. The

48 ASF families occupy a little more than one fifth of the

total reserve area. These inhabitants practice rubber

extraction and Brazil nut gathering, along with basic

subsistence agriculture, animal husbandry, fishing, and

hunting.

Since its creation in 1988, the ASF has been managed by

the Technological Foundation of Acre State (FUNTAC). From

1989 onward, FUNTAC, supported by the International Tropical

Timber Organization (ITTO) has been conducting studies on

socioeconomic conditions, vegetation, soil, fauna, river

CHAPTER 3
STUDY SITE AND METHODOLOGY

Location of Study Site

The Antimari State Forest (ASF), located 80 km north of

the state's capital Rio Branco, was part of the former

Arapixi, Pacatuba, and Mapinguari rubber estates. The

forest extends over an area of 66,168 ha (FUNTAC, 1991a)

between the meridians 68000'19" and 68021'45" W, and the

parallels 09010' and 09031'S (Figure 3.1). In the flat to

lightly undulating terrain lie the settlements of 48

families (300 people), who live and work there, and another

25 families (150 people) within the area of influence. The

48 ASF families occupy a little more than one fifth of the

total reserve area. These inhabitants practice rubber

extraction and Brazil nut gathering, along with basic

subsistence agriculture, animal husbandry, fishing, and

hunting.

Since its creation in 1988, the ASF has been managed by

the Technological Foundation of Acre State (FUNTAC). From

1989 onward, FUNTAC, supported by the International Tropical

Timber Organization (ITTO) has been conducting studies on

socioeconomic conditions, vegetation, soil, fauna, river

........ is Amazonas 46800'

i"

Ac -
aJ

~ ^ f \. A

Sntimari \ '.pu

6L ,
i \ '

I"
LV.

Figure 3.1 Location of the Antimari State Forest (ASF).

a0 2 Km
+9030 -
a60000.

basins, production systems, ethnobotany, and economic

botany. Assessments of the timber industry in Rio Branco

have been extensive. These studies seek alternative

management plans for the ASF. However, the social

organization needed to establish conditions required for

alternative work, has not been developed in the ASF.

Socioeconomic Aspects

Colonization of the state of Acre occurred at the end

of the XIX century, when the extraction of latex from native

rubber was an important economic activity. Affected by the

rubber industry bust, rubber tappers shifted their efforts

to the collection and selling of other products, Brazil nut

being the most important. These two complementary

activities constituted the base of the state's economy for

several decades, contributing an important percentage of the

primary sector to the state's economy (FUNTAC, 1991a).

Policies for the Amazon occupation have undergone

considerable change since 1964. The post-war subsidies for

extractive activities were cancelled, while agricultural

production was given an incentive. As a consequence, the

majority of rubber estate owners were forced to sell their

lands because they had incurred so many debts. Rubber

tappers, however, remained in the area performing extractive

activities in their production units (colocag5es) (FUNTAC,

1991a).

This situation is not restricted to the ASF. Old

rubber estate owners, for example, did not even sell their

land but simply abandoned the area. One of the tasks of

FUNTAC and also of the Institute for Colonization and

Agrarian Reform (INCRA), is to make the old owners sell

their land to the state in order to establish a

comprehensive development project based on the sustainable

use of the forest resources.

The largest number of people to live in the ASF was 250

families. The census conducted by FUNTAC in 1991, found 62

families in the ASF and 32 in its surroundings. From this

group, FUNTAC interviewed 80 families (53 from the ASF, and

27 from its surroundings) totalling 441 persons (107 adults

and 334 individuals younger than 20 years). Thirty-seven

percent were males and 62.5% were females. Seventy-two

percent of the total were originally from the state of Acre.

Ninety percent of these owners of the colocagao or

"posseiros" interviewed were illiterate. Of the families

that settled in the ASF, 20 resided along the banks of the

river and 42 in the interior (FUNTAC, 1990a).

FUNTAC (1991b) found three forms of migration in the

ASF: from one colocagao to another, from one rubber estate

to another, and from the ASF to the city. Causes cited for

these migrations are low productivity of rubber trees, game

depletion, large distance from good fishing sites, absence

of Brazil nut trees, bad relations with the patron, and

relatives living too far from each other.

The socioeconomic order is three-tiered, consisting in

ascending order of the rubber tapper, the "marreteiro,"I and

the patron. The patron at the top of the social pyramid

centralizes the production of a group of colocacges and

interchanges manufactured products with the rubber tappers

based on the aviamento system. Below the rubber tapper

there may be the "meeiro." This is a rubber tapper without

a colocagao of his own who works primarily in the extraction

of rubber for another rubber tapper who owns a colocagco.

The production arrangements are ruled by the levels

described above. In the survey performed by FUNTAC there

were 18 buyers (patrons and "marreteiros"), two of whom

monopolized half the production. Although some rubber

tappers owned boats to transport their products, they

preferred selling their products to the nearest patron

instead of trying to find a better price in a place where

transactions are conducted in cash. This is largely due to

their illiteracy; they favor bartering their products for

available goods as opposed to dealing with cash

transactions. Basically, patrons and "marreteiros" purchase

only rubber and Brazil nuts from the colocagoes.

A person that usually does not live in the ASF but owns a small
watercraft to interchange commercial products for forest products

71

According to their economic activities, rubber tappers

can be divided into three groups: rubber and Brazil nut

collectors, collectors who also are small scale agricultural

producers, and former collectors dedicated to small scale

subsistence farming (FUNTAC, 1991b). However, at the time

of this study only one family belonged to the last group.

Families normally work alone. With the exception of

labor pooling for land clearing, there are few other family

interchanges. There is little community organization,

however, some family members are interested in establishing

this link.

Production Systems

In 1990 the total production of rubber in the ASF was

40 tons, with an average of 753 kg per colocacao. All the

rubber produced is coagulated, pressed, and sold in blocks

that average 50 to 60 kg. Approximately 17% of each

family's hand labor is involved in this task. Considering

the fact that there are 544 trails for rubber extraction

(with an average of 114 trees) of which 442 have been used,

FUNTAC (1991b) concluded that 13.6% of the existing rubber

trees are utilized.

The production of Brazil nuts in the ASF was 72 tons

(1990). In contrast to rubber trees, Brazil nut trees do

not appear homogeneously spread throughout the ASF. There

are colocagces with no Brazil nut trees, and others where

72

production is strictly for family consumption. Based on 13

colocacges, FUNTAC (1991b) estimated an average production

of 100 cans or 1300 kg per colocagao.

Taking the prices from Rio Branco ($0.60 per kilogram

rubber in block and $0.45 per kilogram unprocessed Brazil

nuts), FUNTAC (1991b) estimated an average annual gross

income of $980 per family.

Agricultural production is practiced on an average of

3.63 tarefas (0.9 hectares) per colocagao. In this area

rice with maize, beans, and manioc are planted in sequence,

for a maximum period of three or four years. The soil is

cleared by burning, leaving 1.25 hectares as secondary

succession forest per colocagao.

Nine families raised cattle in 1991, with an average of

1.88 hectares (7.5 tarefas) per colocagao and an average of

6 animals (1.25 tarefas per head). Some fruit species such

as assai, pataua, and bacaba are collected and consumed in

the area, as are the products of hunting and fishing. Men

and women spend an average of four hours each time they go

fishing, with males spending less time than females. The

use of bait and line yields about two meals per fishing

period.

The average time spent hunting is five hours per

hunting period. Hunting periods occur about 8 days per

month. The products of hunting and fishing make up 46.5% of

the food consumed: fish (15.8%), deer (Mazama spp) (13%),

73

peccary (Tajassu tajacu) (5.1%), and others. The ingestion

of protein per kilogram of body weight among those residing

in the ASF was estimated to be 0.73 g per hunting day

(FUNTAC, 1991a).

Housekeeping and child care is almost entirely a female

responsibility. There is no information from ASF, but a

study conducted in the Chico Mendes reserve showed that

these activities absorb 10% of labor demands in one

colocagco (CNS, 1992).

Fauna

Studies of wild animal densities and their diets were

performed by FUNTAC between June 1990 and April 1991. An

evaluation of the densities of 21 animal species showed most

hunted species in the ASF had lower population densities

than those in less populated Amazon areas where similar

studies have been conducted (FUNTAC, 1991c). In addition,

samples of 311 stomachs collected showed that some plant

species consumed were of commercial value (Table 3.1).

Vegetation

The ASF contains an average of 115.5 trees per hectare,

with a mean basal area of 15.23 meters per square hectare.

The volume, including the bark, is 129 cubic meters per

hectare (FUNTAC 1990). Based on the different forest types

surveyed, three management areas were defined (Table 3.2):

73

peccary (Tajassu tajacu) (5.1%), and others. The ingestion

of protein per kilogram of body weight among those residing

in the ASF was estimated to be 0.73 g per hunting day

(FUNTAC, 1991a).

Housekeeping and child care is almost entirely a female

responsibility. There is no information from ASF, but a

study conducted in the Chico Mendes reserve showed that

these activities absorb 10% of labor demands in one

colocagco (CNS, 1992).

Fauna

Studies of wild animal densities and their diets were

performed by FUNTAC between June 1990 and April 1991. An

evaluation of the densities of 21 animal species showed most

hunted species in the ASF had lower population densities

than those in less populated Amazon areas where similar

studies have been conducted (FUNTAC, 1991c). In addition,

samples of 311 stomachs collected showed that some plant

species consumed were of commercial value (Table 3.1).

Vegetation

The ASF contains an average of 115.5 trees per hectare,

with a mean basal area of 15.23 meters per square hectare.

The volume, including the bark, is 129 cubic meters per

hectare (FUNTAC 1990). Based on the different forest types

surveyed, three management areas were defined (Table 3.2):

Table 3.1 Ingested plant residues found in the stomachs of
about 300 animals from the ASF.

Vegetal Part of plant Animal species Freq.
species consumed common name
common name
Assai fruit aracua 1
jacu 2
nambu preta 1
veado capoeiro 3
Bacaba fruit guariba 1
Brazil nut fruit cutia 1
macaco prego 1
Copaiba fruit cutia 2
paca 2
porquinho 1
veado capoeiro 1
Pau d'arco flower guariba 2
Jatoba fruit cutiara 1
manite flower, leave jabuti 2
Pau marfim fruit porquinho 1
Murmuru fruit porquinho 8
Pataua fruit porquinho 1
Pau sangue fruit nambu galinha 1
Sumauma seeds nambu galinha 1
Source: FUNTAC, 1991c

Table 3.2 ASF management areas, based on forest types.

Region Forest Type Area (ha)
1 Dense Forest 12,597.17
2 Dense Floodplain 17,904.05
Forest
3 Open Forest with 34,751.69
bamboo
TOTAL 65,252.91
Source: FUNTAC, 1992

(1) dense forest, (2) dense floodplain forest, and (3) open

forest with bamboo (Figure 3.2).

In the first region 216 tree species were counted, 311

in the second, and 232 in the third. In the first two

69-01
+ -W

V- Ol Dense forest
\ I Dense floodplain forest
\ ] l Open forest with bamboo
Fi 9u'30' 3 Ac 2

Figure 3.2 Management regions,-and studied households in the ASF, Acre, Brazil. Ln

Table 3.3 Main parameters determining the commercial
potential for wood in the ASF, by management region.

Item Region Region Region
1 2 3
Volume per tree (m-) 1.08 1.04 1.36
Pot. Useful Vol. >= 40 (m3/ha) 97.11 87.99 71.82
Volume with DBH <40 cm (m3/ha) 46.96 49.66 26.43
Non-useful Q.C. (m3/ha) 4.44 8.55 3.17
Total potential volume (m3/ha) 148.51 146.20 101.43
Basal area (m2/ha) 17.29 17.14 11.56
Abundance (# ind./ha) 137.87 140.88 74.81
Number of Species 216.00 311.00 232.00
Potential exploitable vol. 48.68 27.88 27.74
(m3/ha)
Pot.expl. abundance (# 17.77 11.75 9.33
ind./ha)
Source: FUNTAC, 1992

regions, the total volume, as well as the basal area and

abundance of species was higher than the third region (Table

3.3).

Still the volume in the third region approaches the

estimates of the other two in terms of potential wood per

tree with a DBH higher than 40 cm. Also, in terms of

exploitable volume, the volume estimated for the first

region was considerably superior to the second and third

regions.

The economic potential for wood harvest from the three

regions was based on a forest inventory and a bibliographic

survey for the species found. In this manner, 82 species

were selected by FUNTAC. The economic characterization of

these species according to the volume of trees with DBH

higher than 40 cm showed that regions 2 and 3 could produce

similar volumes, but less than region 1 (Table 3.4).

Table 3.4 Timber potential of the 82 tree species selected
according to their potential for timber, by management
region.

Region Area Species Vol/ha Ind/ha Average log
(ha) number (m3) price (US
$/m3)
1 12,597.2 52 48.68 17.77 26.16
2 17,904.1 49 27.79 11.75 24.45
3 34,751.7 46 27.74 9.33 28.24
TOTAL 65,252.9 82 31.80 11.62 26.72
Source: FUNTAC, 1992

A study of costs for timber production was conducted by

FUNTAC. The study was based on a model for mechanized

commercial harvest and rotation schedules of 25 years, and

indicated that the cost of log production fluctuated between

US $28.3 and US $47.3 per cubic meter (Table 3.5). This

means that the price of the product barely covered the cost,

even if the yield had been higher than 25 cubic meters per

hectare.

Table 3.5 Average revenues and costs of different logged
volumes potentially harvested in the ASF.

Average Average Revenue Average Cost
logged
volume(m /ha) (US $/m') (US $/m )
35 28.49 28.31
30 29.58 29.14
25 30.49 30.43
20 31.87 32.32
15 34.16 35.20
10 38.74 39.42
7 44.63 47.30
Source: FUNTAC, 1992

A major percentage of the costs incurred stems from the

construction of access roads. For just one area of 4,000

hectares out of the six projected, the road construction

cost was more than $267,000. Additionally, the transport of

wood to the principal center of production in the ASF would

cost $3.11 per cubic meter, with the transport from there to

Rio Branco (135 km) an additional $14.06 (Table 3.6).

From these figures, FUNTAC concluded that due to the

large distance of the ASF from Rio Branco, the

commercialization of wood logs was not economically

feasible. It would be necessary to process the wood on site

and harvest a volume of 35 cubic meters per hectare if

returns were to justify the investment. In addition, in

Table 3.6 Log production costs, based on different
harvesting intensities in the ASF.

Harvesting intensity
Activity 20 25 30 35
m3/ha m3/ha m3/ha m3/ha
---US $---
100% Inventory 0.70 0.70 0.70 0.70
Permanent roads 3.75 3.00 2.50 2.14
Secondary roads 3.08 2.46 2.00 1.75
Skidding trails 0.28 0.22 0.18 0.16
Log landing 0.94 0.80 0.72 0.65
Felling/cross cutting 0.56 0.56 0.56 0.56
Skidding 3.16 3.16 3.16 3.16
Loading 0.38 0.38 0.38 0.38
SUB TOTAL 12.85 11.28 10.20 9.50
Silvic. treatm. cost 0.42 0.42 0.42 0.42
Dynamic inventory 0.13 0.10 0.08 0.70
SUB TOTAL 13.40 11.81 10.70 9.99
Second. transp. 14.06 14.06 14.06 14.06
(135 Km)
sub total 27.46 25.87 24.76 25.05
Contingency and 4.86 4.56 4.38 4.26
administration (17%)
TOTAL 32.32 30.43 29.14 28.31
Source: FUNTAC, 1992

order to reduce costs, a portable log processing plant to

allow the transport of sawn timber from the ASF to Rio

Branco would have to be installed.

To examine the potential use of other species from the

ASF's floristic composition, a botanical study was conducted

which registered 844 species (FUNTAC, 1990). Considering

their potential use by the rubber tappers, 431 species were

classified in 14 different categories (Table 3.7). However,

a characterization or efficiency test of the considered

species was not performed, with the exception of a test

conducted by the National Service for Industrial Learning

(SENAI) regarding the diversification of species used in

timber and furniture (SENAI, 1994).

Table 3.7 Potential uses of different vegetal life forms in
the ASF.

Life form
Potential use Herb Shrub Liane Palm Bamboo Tree
Man food 2 3 15 36
Game food 2 5 5 19 54
Civil construction 11 64
Ship timber 4
Fishing and hunting tools 1
Utensils 1 11 1 10
Fibers 1 5
Medicaments 11 9 8 2 3
Tannins 2
Ornaments 1
Insect repellents 1
Firewood 34
Others 3 4 18 3 1 28
Source: FUNTAC, 1990

Considering the necessity for (a) species

identification, technological properties determination, and

classification into categories for final use, (b) studies

about "unknown" timber types, and (c) finding substitutes

for the few species currently demanded in Rio Branco2; the

SENAI developed a project for the diversification of wood

species for timber and furniture. The aims were to find

substitutes for some of the species in high demand, and to

plan and execute programs for their market introduction.

From forest inventories conducted in the ASF, SENAI selected

20 species that offered excellent potential for both timber

and furniture use (Table 3.8).

Contrary to the ECOTEC findings (discussed in Chapter

2), a preliminary FUNTAC market analysis seeking potential

NTFPs products, revealed a market for copaiba (Copaifera

multijuga) derivatives such as soap and oil. The oil is

marketed in both bottles and capsules (FUNTAC, 1993). Some

laboratories in S&o Paulo pay up to $5 per liter oil, and

sell it to pharmacies at for $10 per liter. The pharmacies,

in turn, sell the oil in 30 mm bottles at $1.50 per unit.

In S&o Paulo approximately 2,000 liters are sold per year.

In Rio Branco (August 1993) copaiba oil was sold at $0.75

per liter. Also, different copaiba soaps are sold for $2.50

per unit.

2 Only 3 species in the furniture sector and about 9 in the timber
sector comprise about 90% of Acre's timber industries, causing a rapid
depletion of their tree sources.

MILLISECOND	CLASS.METHOD	MESSAGE
0	sobekcm_page_globals.constructor
0	sobekcm_page_globals.constructor	Application State validated or built
0	sobekcm_database.verify_item_lookup_object
0	sobekcm_page_globals.constructor	Navigation Object created from URI query string
0	sobekcm_database.verify_item_lookup_object
0	sobekcm_page_globals.display_item	Retrieving item or group information
0	sobekcm_page_globals.get_entire_collection_hierarchy	Retrieving hierarchy information
0	sobekcm_assistant.get_entire_collection_hierarchy
0	cached_data_manager.retrieve_item_aggregation
0	cached_data_manager.retrieve_item_aggregation	Found item aggregation on local cache
0	item_aggregation_builder.get_item_aggregation	Found 'all' item aggregation in cache
0	system.web.ui.page.page_load (ufdc.page_load)
0	sobekcm_page_globals.constructor.on_page_load
0	html_echo_mainwriter.add_style_references	Adding style references to HTML
0	html_echo_mainwriter.add_text_to_page	Reading the text from the file and echoing back to the output stream
30	html_echo_mainwriter.add_text_to_page	Finished reading and writing the file