Technical ReportPDF Available
Development without Deforestation
A publication of
The International Policy Centre for Inclusive Growth
United Nations Development Programme August 2014 No. 29
Policy in Focus
The International Policy Centre for Inclusive Growth (IPC-IG) is a joint project
between the United Nations and Brazil to promote South-South learning on social
policies. It specialises in research-based policy recommendations on how to reduce
poverty and inequality as well as boost inclusive development. The IPC-IG is linked
to the UNDP Brazil Country Office, the Secretariat of Strategic Affairs (SAE) and the
Institute for Applied Economic Research (IPEA) of the Government of Brazil.
Director: Jorge Chediek
Senior Researcher: Diana Oya Sawyer
Policy in Focus is a regular publication of the International Policy Centre for Inclusive
Growth (IPC-IG). This special edition was made in partnership with UNDP Brazil
exploring the work of the UNDP and the Global Environment Facility (GEF). We would
like to recognise the many partners who have contributed to the UNDP Brazil project
implemented in the northwest of Mato Grosso, outlined in this publication, and also
thank the GEF ( for their financial contribution towards that initiative.
Editor-in-Chief: Michael MacLennan,
UNDP/International Policy Centre for Inclusive Growth
Specialist Guest Editors: Carlos Ferreira de Abreu Castro, Coordinator and
Guilherme B. R. Lambais, Consultant, Sustainable Development Unit, UNDP Brazil
Copy Editor: Jon Stacey, The Write Effect Ltd.
Publications Manager: Roberto Astorino
Art and Desktop Publishing: Rosa Maria Banuth and Paula Simone
Cover Art: Genesis, by Daniel Malta
Editor’s note: This special edition presents some of the ongoing discussions about
forest frontier regions of the Amazon and the Himalayas surrounding the themes
of deforestation, degradation and their juxtaposed or complementary relationship
with development. The contributing authors were able to raise some pertinent
questions about the future of economic growth as well as the biodiversity and
communities that exist in such regions today. Simultaneously, drawing heavily
from the experiences of the UNDP project with GEF financing, the collective body
of articles highlights some successful cases of development without deforestation,
providing insight for forest frontier regions around the world.
On behalf of the UNDP IPC-IG I am grateful to UNDP Brazil for their support in the
development of this special edition, in particular to our Specialist Guest Editors
Carlos Ferreira de Abreu Castro and Guilherme Lambais as well as Manoel Salles
for his dedication to the publication of this issue. Finally, I would like to express
my sincerest appreciation to all of the authors for their generous and insightful
contributions, without which this publication simply would not have been possible.
This special edition is dedicated to Jorge Luis Vivan who passed away during the
development of this issue. Our thoughts go out to his friends and family.
International Policy Centre for Inclusive Growth
United Nations Development Programme
SBS, Quadra 1, Bloco J, Ed. BNDES, 13º andar
70076-900 Brasilia, DF - Brazil
Telephone: +55 61 2105 5000
E-mail: URL:
Editorial Office
© 2014 International Policy Centre for Inclusive Growth
United Nations Development Programme
ISSN: 2318-8995
The views expressed in IPC-IG publications are the authors’ and not necessarily
those of the United Nations Development Programme or the Government of Brazil.
Rights and Permissions – All rights reserved. The text and data in this publication
may be reproduced as long as written permission is obtained from IPC-IG and the
source is cited. Reproductions for commercial purposes are forbidden.
The Frontiers of Interdisciplinary Research on
Development without Deforestation
On the Concept of the Frontier
Integrated Conservation and Development
Projects in the Amazon: an Interdisciplinary
Evaluation with Lessons for Forest Frontiers
A Policymix: Conservation and Sustainable
Use of Biodiversity with Poverty
Reduction in the Amazon
Developing Forest-sector and REDD+
Governance: a Multi-stage, Multi-level and
Multi-stakeholder Approach in Nepal
The Necessity of Land Governance:
Sustainable Development in the Amazon
Integrating Livelihoods and Land-use
Change at the Frontiers of Deforestation
Beyond the Panacea: a Critical Assessment
of Instruments of Deforestation Control
Deforestation in the Himalayas:
Myths and Reality
Property Rights, Deforestation
and Violence: Problems for the
Development of the Amazon
Social Policies and Forest Frontiers: the
Consequences for Agricultural
Land-use in the Brazilian Amazon
Sustainable Settlements in the Amazon
Marketing of Agroextractive Products:
Problems and Solutions
Modelling of Deforestation Scenarios
for the Northwest of Mato Grosso
he conservation projects managed by the United Nations Development
Programme (UNDP) in Brazil are underpinned by a strong element of inclusive
local development, consisting of innovative initiatives regarding the sustainable
use of biodiversity. There are many examples of projects in Mangrove, Caatinga
and Cerrado areas, which conciliate the production of goods and environmental services
with the generation of jobs, income and an increase in life quality.
It is always an enormous challenge to assure opportunities to the most vulnerable
populations living in areas characterised by a great expansion of agricultural
commodities production and cattle ranching, while respecting people’s livelihoods
and protecting the environment.
In the Brazilian Amazon this challenge is even greater. We need to adopt agriculture and
forestry systems with higher production and lower environmental impacts in order to
overcome the traditional and prevailing “slash and burn” model. How should biodiversity
conservation projects be implemented in areas of intense conflict involving, among
others, indigenous populations, loggers, poachers, smallholders, cattle ranchers and
agro commodities producers?
For many years, the cities in the Northwestern region of Mato Grosso have been at the top
of the list of municipalities with the highest number of murders per capita and the highest
deforestation rates in Brazil. It was in this region that UNDP and the State Environment
Secretariat developed a project, financed by the Global Environment Facility, which
demonstrated that there are sound alternatives for development in the Amazon region.
Such alternatives could simultaneously conserve natural resources and produce goods
and services, as exemplified by the production of Brazil nuts in indigenous areas and
land reform settlements. The standing forest is generating income which is considerably
higher than the income deriving from the surrounding deforested areas.
As Jorge Luis Vivan—who has recently passed away, and to whom we dedicate this
issue—used to emphasise, agroforestry is another great example of a production model
that generates an income far superior to extensive cattle ranching and other land uses,
which unfortunately hasn’t received the appropriate attention from finance and research
organisations. The lack of solid public policies, credit and technical support geared towards
the sustainable production of timber and non-timber products elicit the conditions
for further deforestation. As evidenced by the articles in this issue of Polic y in Focus,
deforestation by itself greatly increases the land value in forest frontier areas.
The goal of this edition is to stimulate the discussion around these issues and the
associated challenges they present. We wish to better understand the obstacles for a
development process that respects human and biological diversity.
The extremely intense transformation and degradation of the Amazon landscape occurs at
a blinding speed, requiring coordinated long-term investments from technical cooperation
agencies, development banks and finance funds, in order to oppose the driving forces of
tropical deforestation and socio-cultural degradation.
by Carlos Ferreira
de Abreu Castro
The Frontiers of Interdisciplinary Research
on Development without Deforestation
by Guilherme B. R. Lambais
Establishing the foundations of human
development without deforestation is
undeniably one of the backbones for
safeguarding the economic system and
the preservation of the species. Today it is
known that life on Earth is a complex system
and, henceforth, on the threshold of order
and disorder. If unrestrained deforestation
continues to occur and this threshold is
surpassed, while endangering numerous
species, it will also create new risks to the
environment and the process of evolution
itself, including the continuity of human life.
The Earth can be conceived as having
nine thresholds, which together form the
planetary boundaries in relation to a safe
operating space for humanity. These are
considered to be: (1) climate change;
(2) the rate of biodiversity loss (terrestrial
and marine); (3) the interference with
the nitrogen and phosphorus cycles;
(4) stratospheric ozone depletion; (5) ocean
acidication; (6) global fresh-water use;
(7) the change in land-use; (8) chemical
pollution; and (9) atmospheric aerosol
loading. The boundaries in three systems—
the rate of biodiversity loss, climate
change and the human interference
with the nitrogen cycle—have already
been exceeded (Rockström et al., 2009).
Deforestation is assumed to be one of the
main causes for exceeding these boundaries.
This issue of Policy in Focus contributes
to the growing literature that aims at
analysing human development while
also preserving forests in the Global
South. This means examining how to
create value, conserve and sustainably
use the biodiversity that forest areas
have to oer, and how to establish the
long-term conditions so that forests are
considered to be more valuable standing
rather than cut down and commodied.
The frontier areas of forests are one of
the main loci of interaction between
biophysical and human elements, which,
in turn, determine the rate of landscape
transformations. Forest frontier regions
are peculiar because they also demarcate
areas of internal expansion of a country,
given that external (political) borders are
usually already well-dened.
It is useful, therefore, to outline the concept
of the frontier. This is an issue that is often
overlooked, despite its importance for
understanding the reality of these areas
and enabling feasible political propositions.
The authors Vitor Fernandes and Bastiaan
Reydon undertake this task in the rst
article of this issue. Despite the considerable
geographic and historical specicity of
this concept, the authors’ analysis takes
into consideration the main theoretical
underpinnings for the understanding of the
frontier concept, which can be applied to
most frontier regions of the world. Dened
broadly, the frontier can be conceived of as
the expansion front in the anthropological
sense of clashing societies (indigenous
populations, small-scale farmers, loggers,
large landowners etc.) combined with
the pioneer zones in the economic sense
of land-use change directed by modern
capitalist structures.
Organising the geographic space in these
areas remains a great challenge, because
of the broad spectrum of issues involved
therein; these forest frontiers are:
the areas of the most biodiversity
in the world;
the home to established indigenous
the main migration route for
poor smallholders in search
of opportunities;
the pioneer zones where international
and domestic market forces reach
out to, in the hope of high prots
originating from natural resources
(mining, agriculture and energy); and
the places where large landowners
set to expand their activities
(cattle-ranching and agriculture).
Following the 1992 Rio Conference, the
Northwest region of the Brazilian state
of Mato Grosso was chosen to be a focal
point for various pilot projects, including
a major project funded by the Global
Environment Facility (GEF) with a set
of activities dealing with alternatives
to deforestation and local sustainable
development under the coordination of
Carlos Castro of UNDP Brazil. The second
article by Jorge Luis Vivan (in memoriam)
et al. analyses the UNDP project (nanced
by a GEF grant) employing primary and
secondary data on biophysical, socio-
economic and institutional parameters on
small farms and land reform settlements.
As one of the main ndings, the authors
state that certain individual small farms
with intensive agroforestry land use
demonstrate up to 63 times the revenue
per hectare than cattle ranching—an
impressive achievement from a socio-
economic standpoint. However, projects
that best demonstrated forest preservation
across the landscape did so through
support for cooperative infrastructure
and institutional arrangements, including
certication, for forest extractive and
agroforestry commodity supply chains.
Rather than designing incentives around
individual farm properties, such supply
chain interventions were structured over
longer time periods and involved an entire
land reform settlement as well as several
indigenous territories in the region.
Similarly, Peter May et al. turn to the same
projects as the previous article but do so
from an analytical policymix perspective.
This analytical framework sets out to
examine the existing policies in an
integrated and dynamic manner, because
one level of policy intermingles with
another, both horizontally (economic,
agricultural and environmental policies)
and vertically (the municipal, state and
federal government levels as well as
the project and landscape levels). The
authors highlight the positive eects
of the projects at the landscape and
The International Policy Centre for Inclusive Growth
Policy in Focus 5
municipal levels, such as the reduction
of deforestation and the establishment
of community governance, although
they point out that further work still
needs to be done in this eld.
Analyses demonstrate that adequate
governance is essential for the
establishment of the possibility
of development without deforestation.
The fourth article, by Frederico Lopéz-
Casero et al., provides an innovative
approach for developing the forestry-
sector and REDD+ (Reducing Emissions
from Deforestation and Forest
Degradation) governance. The authors
explore, through action research in Nepal,
how the development of a multi-stage,
multi-stakeholder and multi-level process
can ensure eective and legitimate
governance for forest carbon emissions
trading. Fundamental for proper REDD+
governance and for curbing deforestation
is land governance itself. Bastiaan
Reydon and Vitor Fernandes, in the fth
article, clearly demonstrate why land
governance is essential for the sustainable
development of the Amazon: their
research shows that, on average, cleared
areas are worth four times more than
they would be worth as standing forests.
This indicates that land governance should
be one of the rst measures to be taken
when implementing any type of policies
that aim at preventing deforestation,
because, rst and foremost, any possibility
of proting illegally from cleared land
(through logging and the speculation
of land assets) must be eliminated.
In a related discussion, Aldicir Scariot
describes the international debate about
policies opposing land sparing—that
is, the total protection of some areas,
prohibiting any form of land use—and
land sharing, which is the sharing of
activities, including protection, in a given
area. Scariot takes the position that total
land sparing is not feasible in areas of
indigenous populations because this type
of policy would penalise them the most.
Therefore, it is desirable for both policies
to be integrated into what the author
proposes as a mixed strategy.
It should be evident by now that there
are many instruments for achieving
human development without resorting
to deforestation. The seventh article—by
Raoni Rajão et al.—provides a critical
assessment of the instruments for
controlling deforestation. The objective
is to go beyond the ‘panacea’ of the
search for an ‘optimal’ instrument.
Most likely there is no such thing as an
optimal policy regarding this issue, and
we are best served by engaging with
an informed mix of various types of
policies, described by the authors.
In the eighth article, Jean-Marie Baland
and Dilip Mookherjee take the issue of
deforestation back to the Himalayas to
the forest region that stretches across
Nepal and India. The authors provide a
critical assessment of the issue therein,
providing a review of theory and
evidence, using an extensive and detailed
variety of micro-level data sets, in order
to evaluate the causes and the rate of
forest degradation in the region, thus
providing short- and long-term policy
Some socio-economic issues pertaining
to frontier regions are dealt with from
the ninth to the twelfth article. The ninth
article by André Sant’Anna and Carlos
Young, analyses the age-old problem of
the interrelationship between property
rights, deforestation and violence.
The authors, through rigorous analysis,
point to the immediate necessity of
dealing with the issue of property rights
in order to curb both deforestation and
violence at the frontier regions.
The tenth article, by Gabriel Lui, describes
the consequences of social policies,
mainly Bolsa Familia and social security,
for decisions regarding land use in
the Amazon. The author nds evidence
that social policies contribute towards
controlling deforestation by changing
such decisions.
In the eleventh article, Mauro Soave Jr et
al. take on the defence of an alternative
social policy: the establishment of
sustainable settlements in the Amazon
region. The authors argue that since most
of the new settlements in Brazil today are
in the Amazon region, evidence points
to the necessity of curbing deforestation
within the settlements themselves.
The authors thus propose establishing a
model of low carbon agriculture for such
settlements. Furthermore, the twelfth
article by Donald Sawyer introduces an
This issue of Policy
in Focus contributes to
the growing literature that
aims at analysing human
development while also
preserving forests in the
Global South.
This means examining
how to create value,
conserve and sustainably
use the biodiversity that
forest areas have to offer,
and how to establish the
long-term conditions so
that forests are considered
to be more valuable
standing rather than cut
down and commodied.
issue that is important for all other policy
recommendations set forth in this issue—
that is, the marketing of agro-extractive
products, delineating its problems and
some proposed solutions.
Finally, the thirteenth article sees Britaldo
Soares-Filho and Raoni Rajão return
with their modelling of deforestation
scenarios for the Northwest region of
Mato Grosso. The authors compare
three scenarios: business-as-usual,
historical tendency and with the
establishment of governance. As a result,
they demonstrate that if improved
governance is not urgently adopted,
the region will endure severe losses.
We hope that this issue of Policy in Focus
can help to further discussions around
the halting of deforestation, while also
contributing to human development and
to the improvement of public policies, in
order to prevent any worst-case scenarios
from taking place worldwide.
Rockström, J., et al. (2009). ‘A safe operating
space for humanity’, Nature, 461 (7263), 472–475.
On the Concept of the Frontier
by Vitor Bukvar Fernandes1 and Bastiaan Philip Reydon1
Brazil is one of the only countries in the
world that still has an open forest frontier
within its borders. Concerns about the
advancement of the domestic economic
frontier recur with increasing intensity,
due to a wide range of issues among the
players involved. Some issues at hand
include land democratisation, private
appropriation of public lands, theft of
indigenous lands and rural conicts, as
well as how productive agricultural
activities have been expanding over
areas of signicant native vegetation.
In this context, we shall attempt to convey
the ideas of thinkers who have delved
into the subject of frontiers, juxtaposing
them to advance the understanding
of what might be called an internal
frontier movement in Brazil.
It is interesting to start by exploring
the work of Frederick Jackson Turner
(1861–1932) and his hypothesis that
held the existence of a frontier as an
explanation for the type of society that
ourished in the United States. According
to Machado (1992), Turner endeavoured
to explain the genesis of the United States
nation as an autonomous process, based
on the hypothesis that the interaction
between an environment with exceptional
geographical features and the collective
eort of small independent land owners
was responsible for the development
of democratic political institutions in
the country. The availability of “free land”
meant that there was land available that
could be appropriated and transformed
into private property. Furthermore, in
the case of western expansion across the
continent, the frontier would have acted
as a “safety valve” for the impoverished
populations coming from the eastern
United States and Europe, who could
acquire property rights over land and
become independent.
The two caveats to which Machado
draws attention are that the lands
were not empty—as they were
indigenous territories—and that access
to land in the Western United States was
not really free. This is because, despite
the Homestead Act,2 settlement and
transport route advancements actually
served as the regulators of the private
appropriation of land.
José de Souza Martins also opposes the
uncritical appropriation of North American
ideas by stating that, “precisely because
he omitted the struggle for land and
the invasion of indigenous territories in
their own society, Turner is certainly not
the best reference when thinking about
complicated frontier conicts” (2012).
However, the application of these concepts
in Brazil was far from systematic. First, this
was because the continued expansion of
the coee industry in the Southeast region
of the country in the early 20th century
drew the attention of domestic and foreign
social scientists, given the specicity of this
expansion vis-à-vis prior settlement waves
in rural areas. Second, this was also due to
a particular combination of circumstances
that favoured the permanence of
European social scientists in the country
for relatively long periods of time—
especially the geographers Pierre Monbeig
and Leo Waibel, who made signicant
contributions to the conceptualisation
of pioneer zones and frontiers in Brazil.
Leo Waibel (1888–1951) was a geographer
educated in Germany who specialised
in agrarian geography in the tropical
Americas. Based on Turner’s ideas, that
frontiers, in the economic sense, constitute
zones of dierent sizes located between
virgin forests and established populated
regions—pioneer zones—he stated that
the concepts of ‘frontier’ and ‘pioneer’ must
be redened. From this perspective, the
cultivation of land is what constitutes the
economic foundation of pioneer zones, not
extensive livestock activities (Waibel, 1979).
In his comparative studies between the
March to the West in the United States
and the possibility of identifying a March
to the West in Brazil, Waibel sought to
highlight the dierences between the
Brazilian settlement march that occurred
up until 1950. Unlike what happened in
the United States, Brazil’s march expanded
linearly or in cores—without forming
a continuous band—and established
itinerant agriculture as a permanent
phenomenon. Waibel focused on the
itinerant nature of Brazilian agriculture—
since an agricultural system characterised
by non-intensive farming techniques
makes it easier to sell land—and on
the criticism of Brazil’s propensity
to dispersed settlement expansions,
based on unproductive estates and the
predominance of land speculators.
When faced with coee expansion in
the southeast, Pierre Monbeig
(1908–1987), a French geographer,
noted that the settlement followed a
modern conception of colonisation.
This “pioneer coee march” led to the
construction of roads and the emergence
of small urban centres and trade systems.
The practice of urban and rural land
subdivision set this pioneer movement
apart from earlier forms of occupation.
This means that it was irrelevant
whether an area had been previously
appropriated or whether it was inhabited.
In short, “the dynamism of pioneer zones
was the product of a new type of funding,
organisation and scale of enterprises,
composing a modern capitalist economy
[...] Nevertheless, it was not enough to
assert that the ‘conquest of new lands’
meant the integration of rural areas into
a capitalist-style economy [...] the notion
of land ownership is the distinguishing
feature of pioneer zones” (Machado, 1992).
Monbeig observed that large pastoral
estates in Brazil had already been
established in the past on vast tracts
of land within the country’s interior,
many with legal land titles. These lands,
however, had no market value when the
activities took place on grasslands and
savannahs, with very few changes made
to the ‘natural’ landscape. However, in
pioneer zones, where agriculture was the
main activity, expansion took place at the
expense of capital investments required
for deforestation and soil preparation—
The International Policy Centre for Inclusive Growth
Policy in Focus 7
Photo: UNDP Brazil. Coee plantation, next to the municipality of Aripuana, Northwestern Mato Grosso.
i.e., the work that changed the natural
landscape was the same as the work
that assigned value to the land and,
therefore, to the property.
Thus, in describing the ‘pioneer march’
to expand the frontiers of the coee
industry, Monbeig emphasised precisely
the big leap in land prices triggered by
the arrival of this crop, allowing for intense
land speculation. So intense was the
trade in land in the southeast at the time,
that “it was concluded that colonisation
was a secondary matter to the desire
to speculate” (Machado, 1992). This was
only possible because the dynamics of
the coee industry expansion relied on
the extension of cultivated areas, thus
successively displacing the agricultural
frontier with few directly productive
investments. It follows that access to
land and land concentration constituted
a ‘condition’—and not a secondary
aspect—of coee production.
Therefore, as pioneer zones expanded,
the price of land increased, generating
signicant speculation and the creation
of a land market controlled by large
landowners, coee traders and real estate
companies. The control of land access
through prices contributed to the creation
of a ‘free’ labour market and an excess of
labour supply, resulting in only a relative
availability of land in these pioneer areas.
Similarly, one may consider the coee
expansion in the southeast at the
beginning of the 20th century in relation
to the recent expansion of agriculture in
the Amazon region, mostly in the states
of Mato Grosso, Para and Piaui. This
connection is made possible by the notion
of ‘possession’, initially adapted to mobile,
predatory and rudimentary agriculture,
but which gradually became the main
form of land ownership, according to Ligia
Silva Osório (2008). Later, it also becomes
consistent with the explanation of the
determinants of Amazonian deforestation
provided by Reydon (2011), which details
the sequence of the occupation of virgin
land (private or public), the extraction
of timber, the beginning of livestock
husbandry and, nally, the development
of a more modern form of agriculture.
This development process can generate
income and legitimise the occupation
by new owners in the short term,
with almost no resources required.
Anthropologists understand the notion of
frontier dierently. The so-called ‘expansion
fronts’ represent frontiers characterised by
“the displacement of the civilised population
and of economic activities regulated, in
some way, by the market [...]. As suggested
by Darcy Ribeiro, [...] they constitute the
frontiers of civilisation” (Martins, 2012). The
contrast between the pioneer zones and
the anthropological concept of expansion
front is most evident in Os Índios e a
Civilização by Darcy Ribeiro: “here, the land
itself has no value [...] As such, the issue
of ensuring legal ownership of the land
does not exist [...] And this dominion does
not take the form of land ownership,
except when it does so accidentally”
(1977 apud Martins, 2012).
According to Martins, there are two ways
of interpreting frontiers: the ‘expansion
front’ expresses a conception of territorial
occupation held by those who use
indigenous peoples as a reference, while
the ‘pioneer zone’ takes no account of
indigenous peoples and focuses on
businessmen, farmers, merchants,
modern small farmers, and entrepreneurs.
In this sense, Monbeig, for example, denes
the indigenous peoples encountered (and
massacred) due to the expansion of the
pioneer zone in western Sao Paulo
as the precursors of that same frontier,
“as if they were there only temporarily,
waiting for the civilisation that would
bring about their end” (Martins, 2012).
From Martins’ perspective, the mismatch
between the observations of geographers
(and economists) and those of
anthropologists is the result of observations
made in unequal social locations. However,
the mismatch of perspectives is, in this case,
an essential expression of the contradictory
diversity of frontiers, more than a result of
the multiple viewpoints about the concept
of the frontier.
Martins (2012) considers the dierent
historical time-frames of frontiers
as pioneer zones, highlighting the
importance of urbanisation, modern
means of communication, economic
enterprises, the modern mentality and,
especially, the conversion of land into a
commodity. In terms of the expansion
front, the focal point is the design of
the displacement fronts of a Western
population to tribal lands, “when
Frontiers can be
loosely dened as the
synthesis of that which is
dened by anthropologists
as ‘expansion fronts’
and, by economists as
‘pioneer zones’, directly
coordinated by the
capitalist mode
of production.
Integrated Conservation and Development
Projects in the Amazon: an Interdisciplinary
Evaluation with Lessons for Forest Frontiers
by Jorge Luis Vivan (in memoriam),1 Rob Davenport,2 Peter. H. May,3 Paulo César Nunes4 and Cornelius Prins5
As one of the world’s most active forest
frontiers, Northwest Mato Grosso (NW
MT) in Brazil has been a focal point for
various pilot projects for conservation and
sustainable use of forests and biodiversity
following the Rio 1992 conference. These
included the Pilot Programme for the
Protection of Tropical Forests in Brazil
(PPG7) and a major project funded by
the Global Environment Facility (GEF)
and implemented by the United Nations
Development Programme (UNDP).6
These projects incorporated support for
agro-environmental measures, focused
1. Institute of Economics, University of
Campinas, Brazil.
2. The Homestead Act is an agrarian legislation
that was signed by Abraham Lincoln in 1862
within the eort to occupy the American West.
In exchange for a symbolic monetary reward,
those that ventured west would receive plots
of federal land (26.3 hectares) that were already
demarked. This exibility never existed in Brazil.
Even to this day the government of Brazil still
does not control a large share of its public lands.
anthropologists speak of an expansion
front, they do it basically to save on words
when dening what Indians face” (Martins,
2012). An interesting denition is the one
by Roberto Cardoso de Oliveira, cited
by Martins, which denes an expansion
front as a “contact situation, that is, the
methodological assumption of totality,
as is characteristic of dialectical tradition”
(ibid.). It is at this moment of ‘inter-ethnic
friction’ that the expansion front becomes
a point of contradiction and conict.
Here, the form of capital expansion, is an
expansion of the trade and commerce
network from where money is usually
absent, appearing only, still according to
Martins, as a “nominal reference refereed
by those who hold personal power and
have control over material resources, in
their relationships with those who exploit
Indians or peasants” (2012).
Such a market operates through village
merchants, in a way that is monopolistic
and mediated by “violent relations of
personal domination”(ibid), be it in the
trade of products or in labour relations
(characteristically, bondage or peonage).
As such, frontiers can be loosely dened
as the synthesis of that which is dened
by anthropologists as ‘expansion fronts’
and, by economists as ‘pioneer zones’,
directly coordinated by the capitalist
mode of production.
It is in this middle ground that we nd the
diversication of dierent and interacting
historical moments: peasants engaged
in surplus agriculture, prosperous
small farmers, rural entrepreneurs,
integrated and non-integrated
indigenous peoples and even gunmen
in service of big landowners.
In conclusion, we have seen that the
analysis of the frontier movement is
complex and can be easily permeated
by prejudice and distortions that are
extremely harmful, from a scientic
analysis perspective, given the
potential reductionism they may incur.
Even though this article encompassed
only some of the many studies on the topic
of land frontiers, we have nonetheless
attempted to elucidate important ways
of dening what a frontier is, to provide
a better understanding of the denition
of frontiers in Brazil and elsewhere, as
well as its associated methodological
Machado, L. (1992). ‘A fronteira agrícola
na Amazônia brasileira’, Revista Brasileira
de Geograa, Vol. 54, No. 2.
Martins, J. (2012). Fronteira: a degradação do
Outro nos conns do humano, 2nd edition.
Sao Paulo, Contexto.
Monbeig, P. (1945). ‘A zona pioneira do Norte-
Parana’, Boletim geográco, Vol. 7, No. 78.
Reydon, B. (2011). ‘O desmatamento da oresta
amazônica: causas e soluções’, Economia Verde,
No. 8, June.
Silva, L. (2008). Terras devolutas e latifúndio, 2nd
edition. Campinas, Editora Unicamp.
Waibel, L. (1979). Capítulos de geograa tropical e
do Brasil, 2nd edition. Rio de Janeiro, IBGE.
on mosaics of agro-ecosystems consisting
of private forests involving communities
living in buer zones or inside protected
areas. These agro-environmental measures
comprised technical assistance for soil
and water conservation, restoration
of degraded areas, tree planting and
agroforestry systems, appropriate
technology, non-timber forest products
and low-impact forest management.
Such a mix of interventions ts the
mould of Integrated Conservation and
Development Projects (ICDPs). The ICDP
approach has been criticised over the past
decade, often due to a lack of empirical
evidence regarding its biophysical,
socio-economic and institutional impacts
(Minang and Noordwijk, 2013; cf. Ferraro
and Kiss, 2002). With due consideration
of this criticism, and a unique hands-on
opportunity, we decided to conduct
an interdisciplinary evaluation of the
outcomes of ICDP pilots in NW MT.
Our case study employed primary and
secondary data on biophysical, socio-
economic and institutional parameters on
small farms and land reform settlements
The International Policy Centre for Inclusive Growth
Policy in Focus 9
Source: Authors’ elaboration. * Synthesis of economic indicators estimated for 55 farms of 4 to 250 hectares
evaluated in the municipalities of Juína and Cotriguaçú, Mato Grosso, Brazil, 2010.
Notes: AFS = Agroforestry Systems; NTFP = non-timber forest product; USD/PD = return in USD per person
per day; Hectare/PYL = area in hectares of the system manageable by one year of labour per person.
Labour is not included in costs, as farmers do not habitually hire an external workforce; n.i. = no indicator.
Northwest Mato
Grosso (NW MT) in Brazil
has been a focal point
for various pilot projects
for conservation and
sustainable use of forests
and biodiversity following
the Rio 1992 conference.
in NW MT from 1995 to 2012 in the
municipalities of Juína, Juruena and
Cotriguaçú. We used databases generated
from the three ICDP projects carried
out during this period, as well as from
eldwork conducted from 2012–2013.
First, to assess biophysical impacts,
we contrasted carbon stocks and
sequestration rates with indexes of plant
diversity in agroforestry systems. We
assumed that the plant diversity indexes
constituted an important proxy for habitat
quality and biodiversity conditions
(DeClerck and Salinas, 2011; Laurance and
Vasconcelos, 2004). Second, for eects on
land use of individual farms, Vivan (2010)
compiled information on 62 farms in Juína
and Cotriguaçú in the form of farm-level
data and maps. Third, for socio-economic
performance data was retrieved from a set
of 55 farms ranging from 4 to 250 hectares
that were evaluated for land use in Juína
and Cotriguaçú by Vivan (2010) as well as
in Juruena by Nunes and Rugnitz (2011).
Fourth, a landscape scale analysis of
deforestation dynamics was conducted
for three land reform settlements in
which ICDP projects could potentially
demonstrate impacts, depending on the
level of settler engagement (INCRA, 2011):
(1) Settlement Project Nova Cotriguaçú of
99,988.5 hectares and 1234 households
settled in 1995; (2) Settlement Project
Iracema in Juína, of 18,120 hectares and
343 households settled in 1996; and
(3) Settlement Project Vale do Amanhecer
in Juruena, of 14,400 hectares and
243 families settled in 1998. For these
settlements, we tracked and compared
Landsat imagery between 1995 and 2011.
Finally, our institutional analysis derived
from household questionnaires and semi-
structured interviews with 29 farmers from
these land reform settlements. Furthermore,
we cross-referenced this data with a group
workshop conducted with settlers of Vale
do Amanhecer, to further understand the
historical evolution of land use decision-
making and institutional arrangements.
Results and discussion
At the individual farm level, biophysical
analyses revealed that agroforestry
systems allowed carbon stocks to increase
at the farm level from 5 to 8 tonnes of
carbon/hectare per year, over three
to thirteen year cycles, for an average
of 2.5 hectares per farm. The analyses
also demonstrated favourable indexes
of tree diversity and potential habitat
connectivity. In terms of land use of
individual farms, participants in ICDPs
demonstrated 13 per cent greater forest
cover than did non-participants.
Considering socio-economic conditions,
we identied a small-farm baseline land
use of mixed beef and dairy cattle, which
uses on average 33 hectares of pasture
while obtaining gross revenues of
USD212/hectare per year or USD6996/year
per household. In contrast, agroforestry
systems induced through ICDPs use on
average 2.3 hectares and obtain gross
revenues of USD4000/hectare per year
for shaded cocoa or a total income of
USD9200/year per household. This is a 56.5
per cent gain, while using less than 7 per
cent of the land area required by mixed
beef and dairy operations (see Table 1).
At the landscape level—i.e. considering the
entire settlement—deforestation in the
Project Vale do Amanhecer did not occur at
the same rate as in the Iracema and Nova
Cotriguaçú projects. All three settlements
were established during the mid- to late
1990s, but in 2011 Vale do Amanhecer
had 22 per cent more remaining forest
than Nova Cotriguaçú and 39 per cent
more than Iracema.
We take particular interest in this result,
arguing that Vale do Amanhecer’s
conserved forest—57 per cent of the
settlement’s total area—is an outcome
of the combination of a series of ICDP
interventions. These interventions
targeted land use regulations, resource
management and economic factors
relevant to cooperative production
and, more specically, the marketing of
non-timber forest products. Furthermore,
the forest area is being managed as a
collective legal reserve forest, in which
7200 hectares are managed for 800
georeferenced Brazil nut trees (Bertholletia
excelsa). Certication for non-timber forest
products is tied to maintenance and
monitoring of the collective legal reserve.
Two settlement-based cooperatives (one
exclusively women-owned and -managed),
established in 2008 and in 2010, process oil,
pasta, our and cookies using Brazil nuts
extracted from the forest reserve. To meet
market demand, nuts are also purchased
from surrounding farmers, ve indigenous
territories and one extractive reserve in the
region. In 2013 the federal National School
Feeding Programme provided USD1,160,000
in advance credit. The advance purchase
allowed these goods to reach 33,000 people
across seven local municipalities.
Institutional eldwork revealed that
these cooperatives emerged in adverse
socio-economic and ecological
circumstances, in which a variety of
factors favoured deforestation throughout
the history of the settlement of Vale
do Amanhecer. These factors included
complex political alliances, widespread use
of re, road building, federal agricultural
credit programmes (PRONAF), an illegal
gold rush invasion of the settlement in
2004, and varying prices of agricultural
commodities versus non-timber
forest products.
The PRONAF7 programme was perceived
as one of the strongest factors
encouraging forest clearing: deforestation
during the operation of PRONAF in 2000–
2001 was nearly four times as high as in
2008. Additionally, ICDPs were forced to
cease operations in Vale do Amanhecer
in 2004 due to violent threats from gold
miners. However, after the federal police
forcibly removed miners from the area
in 2005, activities resumed to encourage
forest conservation.
Settlers observed that pasture-burning
practices were continuous from the start of
the settlement until 2012. However, unlike
the other settlements under consideration,
settlers in Vale do Amanhecer recognised
the social value and legitimacy of
environmental licensing, as this instrument
legalised the cooperatives’ businesses and
established documentation that formally
certied products of the settlement.
These ndings reinforced our hypothesis
that it would be possible to identify ICDP
impacts in NW MT using ner spatial
scales and longer temporal scales. One
hectare of agroforestry system is invisible
from a geographic information system
(GIS) analytical perspective. However, at
the scale of individual farms, one hectare
of agroforestry has the potential to
employ almost 20 times the workforce
and produce 63 times the agricultural
revenues of cattle ranching.
One hectare
of agroforestry has the
potential to employ almost
20 times the workforce
and produce 63 times the
agricultural revenues of
cattle ranching.
The International Policy Centre for Inclusive Growth
Policy in Focus 11
Special thanks are extended to:
André Gonçalves, Pattrikk John Martins,
Luis Henrique da Cunha (consultants); the
AJOPAM team and Ildamir Teixeira in Juína,
in addition to the Policymix Project - funded
by the European Union,9 the National Institute
of Science and Technology on Public Policies,
Strategies and Development (INCT-PPED);
Dr. Carlos Castro (UNDP Brazil); Ms. Rosane
Beatriz Aguiar and Ms. Ana Cristina Balogh
Tripodi (Petrobras); and of course farmers,
leaders of associations, trade unions, and
indigenous reserves.
1. Federal University of Santa Catarina and
Federal Rural University of Rio de Janeiro,
Brazil (in memoriam).
2. University of California, Santa Cruz, USA, and
CATIE (Center for Tropical Agricultural Research
and Education), Costa Rica.
3. Federal Rural University of Rio de Janeiro
and REDES.
4. Juruena Carbon Sink Project,
Mato Grosso, Brazil.
5. CATIE, Costa Rica.
6. A) Pilot Programme for the Protection
of Tropical Forests in Brazil, especially Type
A Demonstration Projects, with expected
impacts in Juína; B) ‘Conservation and
Sustainable Use of Biodiversity in the Frontier
Forests of Northwestern Mato Grosso’, with
expected impacts concentrated in Juruena,
Juína and Cotriguaçú; and C) Juruena
Carbon Sink Project/ADERJUR, with impacts
concentrated in Vale do Amanhecer land
reform settlement.
7. PRONAF - National Family Agriculture
Development Programme, a credit scheme
available for family farmers.
8. Rodrigues et al. (2009) have described
an Amazon rainforest frontier development
pattern in which development indicators are
higher for landscapes transitioning from high
(>70 per cent) to low (<30 per cent) forest
cover. In this pattern, eventually a frontier
development ‘boom’, supported by abundant
resources, shifts to a ‘bust’ of a resource-poor
9. Policymix project <>
funded by the European Commission,
Directorate General for Research.
At the landscape level, signicant
biophysical, economic and institutional
gains were observed in Vale do Amanhecer,
as ICDPs linked cooperatives with
institutions and infrastructure supporting
alternative livelihoods. However, settlers in
Vale do Amanhecer interacted with ICDPs
consistently over the course of at least
eight years, compared to only two to three
years in the Iracema settlement and only
sporadically in Nova Cotriguaçú.
Recommendations and closing remarks
We propose additional research at the
landscape level to assess the impacts and
eectiveness of ICDP interventions, based
on the temporal scale of application.
The synergies produced by the specic
sequencing and combinations of
instruments should also be evaluated. If
farmers are expected to rely on individual
direct payments to access services and
markets, they risk being cut o from
cooperative livelihood and conservation
alternatives promoted by ICDPs and
the long-term processes that these
alternatives involve. Longer temporal
scales may be critical to the eectiveness
of projects and programmes seeking
sustainability for local economies and land
use across the landscape.
As such, the endurance of such ICDP
impacts may depend on whether and
how local stakeholders are able to access
aligned policies and incentives. ICDP
projects in NW MT took place in an
antagonistic politico-economic setting
and in the context of fragile support from
federal and state agencies. Nevertheless,
results from the Vale do Amanhecer Project,
indicate that a combination of instruments
could overcome these limitations, if
applied in an overlapping sequence:
setting priorities by mapping the
potential of the remaining forest;
training and technical assistance;
cooperative social organisation;
legal certication of sustainable
material investments in infrastructure;
market development, credit nancing
and the elaboration of contracts with
surrounding indigenous communities;
contracts with private companies
and the National School
Feeding Programme;
public and political exposure through
national and international recognition
of eectiveness; and
eorts to expand Brazil nut production.
It is by no means inevitable that all
forest frontiers go through the ‘boom
and bust’ transition stages described by
Rodrigues et al. (2009).8 We suggest that
managed forests, like the legal reserves
inside land reform settlements and in
small and medium-sized private farms,
represent an important component for
the sustainability of regional economies.
The projects we analysed may oer
opportunities to mainstream successful
approaches for both collective and
individual incentives, which have proved
eective at the level of land reform
settlements and smallholder farms,
to constitute managed forests as a
continuum between protected areas and
agro-ecosystems (Wiersum, 2004), thus
providing a desired mosaic of sustainable
land use in buer zones around
protected areas of forest frontiers.
DeClerck, F.A.J. and A.M. Salinas (2011).
‘Measuring Biodiversity’ in B. Rapidel, F.A.J.
DeClerck, J. Le Coq and J. Beer (eds), Ecosystem
Services from Agriculture and Agroforestry:
Measurement and Payment. London, Earthscan.
Ferraro, P. and A. Kiss (2002). ‘Direct payments for
biodiversity conservation’, Science, 298:1718–9.
Goncalves, A.L., J. L. Vivan, L.H.H. Cunha and
P.C. Nunes (2009). ‘Evaluation of biological
and economic parameters of Agroforestry
Systems (AFS) being promoted as a strategy of
biodiversity use and conservation in the frontier
forests of northwest Mato Grosso, Brazil’, World
Congress of Agroforestry Book of Abstracts. Nairobi,
UNEP/World Agroforestry Centre: 225–226.
Laurance, W.F. and H.L. Vasconcelos (2004).
‘Ecological eects of habitat fragmentation in the
Tropics’ in G. Schroth, G.A.B. Fonseca, C.A. Harvey,
C. Gascon, H.L. Vasconcelos and A-M.N. Izac
(eds), Agroforestry and Biodiversity Conservation in
Tropical Landscapes. Washington, DC, Island Press.
Minang, P.A. and M. van Noordwijk (2013).
‘Design challenges for achieving reduced
emissions from deforestation and forest
degradation through conservation: Leveraging
multiple paradigms at the tropical forest
margins’, Land Use Policy, Vol. 31: 61–70.
Nunes, P.C. and M.T. Rügnitz (2011). Semeando
Esperança, Colhendo Bens e Serviços Ambientais.
Resultados do Projeto Poço de Carbono
Juruena, 1st edition. Juruena, Associação de
Desenvolvimento Rural de Juruena (ADERJUR)
Projeto Poço de Carbono.
Rodrigues, A., R. Ewers, L. Parry, C. Souza Jr., A.
Veríssimo and A. Balmford (2009). ‘Boom-and-
bust development patterns across the Amazon
deforestation frontier’, Science, 324(5933): 1435–7.
Vivan, J.L. (2010). ‘Desenvolvimento de
instrumentos e parâmetros para recuperação
produtiva de passivo ambiental em
assentamentos e propriedades rurais no
entorno de UCs nos Municípios de Juína e
Cotriguaçú, Noroeste do Estado de Mato Grosso’.
Wiersum, K.F. (2004). ‘Forest gardens as an
“intermediate” land use system in the nature-
culture continuum: characteristics and future
potential’, Agroforestry Systems, 61: 123–134.
Wiersum, K.F. (1994). ‘From natural forest to
tree crops, codomestication of forests and tree
species, an overview’, Netherlands Journal of
Agricultural Science, 15: 425–438.
A Policymix: Conservation
and Sustainable Use of Biodiversity with
Poverty Reduction in the Amazon
by Peter H. May,1 Jorge Luis Vivan,2 João Andrade,3 Maria Fernanda Gebara4 and Pablo del Arco5
This article discusses the pressing issue
of how to achieve the conservation and
sustainable use of biodiversity while also
reducing poverty in the Brazilian Amazon.
The existing mix of policies for biodiversity
conservation and forest protection is
based on traditional command and
control policies and institutional and
economic instruments, such as integrated
conservation and development projects,
a state-based ecological scal transfer
(EFT) instrument, and payment for
environmental services. We question how
best to dene an adequate policymix
to achieve both environmental and
social targets—that is, how to halt the
degradation of natural resources, ensuring
their sustainable use, while also reducing
poverty in the communities where these
natural resources are located.
When addressing an issue from a policymix
perspective, all levels of policy (national,
state and local) should be taken into
consideration. While command and control
policies are established at a national level,
they interact with other policies mostly at
the regional and local levels.
The location of our study lies within a
frontier region of the Amazon, northwest
Mato Grosso. Locally, we are interested
in the biophysical and socio-economic
impacts of the Integrated Conservation
and Development Projects (ICDPs)
that arose out of the United Nations
Development Programme/Global
Environment Facility initiative, specically
those within land reform settlements. At
the regional level, we analyse how the EFT
aects these areas containing the projects,
which are contiguous to indigenous
territories and conservation units.
Will deforestation continue or stabilise
at current levels? What was the role of
the existing policymix in the reduction
of deforestation? Is the prevailing
policymix sucient to achieve
environmental and social targets,
or are new instruments required?
Current policies
for biodiversity conservation
The main policy for biodiversity conservation
adopted in Brazil is primarily associated
with land-use restrictions compiled within
the national Forest Code, dating from 1965.
The public lands of the National System of
Protected Areas (SNUC), the rights of forest
people over indigenous territories and direct
use reserves establish the current framework
for protecting biodiversity and traditional
culture. The federal constitution devolves
responsibility for the licensing of land use
and the permits for forest management
to the individual states.
The federal constitution also permits
the sharing of value-added tax
revenue between state and municipal
authorities. In some states—including
Mato Grosso—such revenues are in part
allocated through the EFT instrument
ICMS-Ecológico (ICMS-E) to compensate
for biodiversity protection.6 The ICMS-E
is, therefore, an economic instrument that
favours biodiversity conservation and
intermunicipal scal balance.
In relation to ICDPs, another policy
currently active in Brazil, it is the
contention of recent observers that this
type of project has often been grounded
in unrealistic expectations (Minang and
Noordwijk, 2012). There is little empirical
evidence of their eectiveness in reducing
deforestation, which prompts a search for
a demonstration of the eectiveness of
ICDPs as well as for alternative and more
targeted economic instruments such
as payment for environmental services
(Ferraro and Kiss, 2002).
Furthermore, restrictions on credit and
government subsidies to municipalities
that are poor performers in reducing
deforestation has motivated local
governments to become more proactive
in preparing for ‘green governance’. Tax
relief for industries of natural products
and voluntary market instruments, such as
a sustainable practice certication and a
soybean moratorium in forest areas, have
apparently been eective in some areas.
Multi-level instrument
interaction in the policymix
The instruments at the disposal of
decision-makers working at the nexus of
agribusiness expansion and biodiversity
conservation in the Amazon are potentially
complementary but reliant on institutional
coordination at all levels to be eective
in halting further adverse and large-scale
land-use change. Notwithstanding, signals
emitted at the federal level regarding the
priority for accelerated national growth,
with the nancing of massive transport and
energy infrastructure expansion deep into
the Amazon, are at odds with voluntary
commitments to reduce greenhouse gas
emissions, which are primarily achieved by
reducing deforestation. Credit policies that
stimulate growth of cattle herds and the
slaughter capacity of facilities at the forest
frontier are equally contradictory, despite
more recent nancing commitments under
the ‘low carbon’ agenda.
Land-use management is best achieved
at a governance level closest to where
the resource is used (Ostrom, 1990). As
Brazilian environmental policy is a shared
responsibility among dierent levels of
government, a rigorous analysis must
pay particular attention to the local,
municipal-level capacities needed to
implement broader strategies, which
rely on inter-sectoral agreement over the
desirable scope of land-use control and
enforcement. Environmental management
on this scale is a very recent assignment
to local governments. Additionally, the
political will to take on the challenge of
The International Policy Centre for Inclusive Growth
Policy in Focus 13
land-use control is reliant on extraordinary
commitments by mayors and municipal
councils, causing signicant dierences
among municipal institutions and their
respective environmental performance.
Some municipal governments are
proactively assuming local commitments
to meet deforestation reduction targets by
improving governance and adopting better
production practices on dierent scales.
Experience with agroenvironmental
measures adopted in southern Europe
oers important parallels (Santos et al.,
2013). However, in the forest frontiers
of northwest Mato Grosso, weak public
institutions, as well as poor spatial
planning and targeting, will need to be
overcome to permit large-scale positive
impacts. In this sense, the use of EFTs to
compensate and stimulate greater local
eorts to protect endangered biodiversity
by using management quality indices
is an experience that is relevant across
continents (Ring et al., 2013).
The eectiveness of the ICMS-E
The Mato Grosso policymix case study
sought to assess the eectiveness of EFTs—
more specically, the role of the ICMS-E
in municipal biodiversity conservation at
the local level—that is, if this instrument
resulted in an increase in the area under
protection. Our rst hypothesis is that
at a municipal level, where tax revenue
discrimination is mandatory by law, the
instrument should have been most
eective at stimulating local
conservation actions.
The national data on the creation of
municipal protected areas were insucient
to test this hypothesis, but in Mato Grosso
there was a clear increase in the number
of municipal protected areas created
immediately after the instrument came
into eect in 2002.
However, due to a resurgence of control
of the state government via agribusiness
lobbying eorts shortly thereafter,
protected areas came under re. This
suggests the need for a detailed appraisal
of the eectiveness of such instruments
at a municipal level, which guided our
site selection and questions raised in
interviews with local government ocials
and stakeholders.
Our concern then focused on mechanisms
adopted locally to better allocate
resources to reect and reinforce the
benets obtained from such conservation.
In Cotriguaçú, for example, although
ICMS-E revenues were signicant, local
environmental council members were
more interested in using these revenues
to keep roads open in the rainy season, to
promote agricultural marketing, than in
better managing protected areas.
When we examined the economic
rationale for this intra-municipal allocation
of ICMS-E funds, we found that the per
hectare revenues from land areas under
protection actually exceeded those
generated by value added from livestock
and timber extraction in some cases (May
et al., 2013). Clearly, the failure to earmark
such tax revenues for environmental
purposes undermines the eectiveness
of the policy. It is thus promising that a
growing number of local governments
have created municipal environmental
funds for which ICMS-E revenues are
earmarked. Such resources could also help
to fortify local environmental governance
through capacity-building among
municipal councillors and authorities.
Equity and legitimacy of the ICMS-E
Since the ICMS-E also includes indigenous
lands in Mato Grosso, our study
hypothesises that enhanced revenues
from this instrument could improve local
governments’ relations with indigenous
groups and fortify their eorts at
environmental management. Our ndings
in this regard appear promising.
In Juína, where 60 per cent of the territory
lies in indigenous lands, the ICMS-E brought
signicant additional revenues, accounting
for 20 per cent of total annual municipal
funds from all sources. As a consequence,
a progressive former mayor proactively
made agreements with indigenous leaders
to transfer 5 per cent of the additional
ICMS-E revenue to projects aiming to
reinforce the integrity of tribal lands and
enhance sustainable incomes. Although the
procedure was armed through municipal
legislation, it required annual renewal and
Source: Authors’ elaboration.
thus was undermined when an opposition
mayor took oce in 2013.
Moreover, surveyed ocials in local
government generally held the opinion
that the ICMS-E represents a top-down
policy instrument, which does not reect
local needs. Additional revenues were,
therefore, diverted to activities that
beneted the population as a whole (road
maintenance, health and education), rather
than to reward private nature conservation
or collective resource management, as had
been done in other states, notably Parana
in southern Brazil. This result reects the
character of general revenue-sharing
attributed to the distribution mechanism,
rather than the use of conditional grants.
The eectiveness of ICDPs
Our study on the eectiveness of a series
of ICDPs focused on comparative studies
of land reform settlements at a landscape
and farm level. These projects encompassed
a combination of technical assistance,
governance capacity-building, provision
of seedlings for agroforestry and forest
restoration, investments in processing
facilities, and the construction of market
channels for non-timber forest products.
Our research methodology consists of
an analysis of three federally sponsored
settlements in three municipalities
in northwest Mato Grosso, including
a quantitative-qualitative analysis of
ecological and economic data on individual
lots, comparing land use, carbon stocks,
enterprise diversity and net incomes among
participants and non-participants in ICDP
activities, both within and outside the
settlements. Furthermore, satellite imagery
was used to examine landscape responses
with the settlement as a landscape unit,
to identify possible spillover eects on
deforestation rates due to the actions
undertaken by ICDP participants.
The integration of environmental objectives
with productive goals in land reform
settlements entails a large measure of
stakeholder participation. In this case,
insight is needed into the producers’
potential gains from individual or collective
property and environmental licensing.
In the case of collective environmental
licensing, the possible cost associated with
vulnerability to increased enforcement
powers also needs to be taken into
consideration. This information is not
easily captured through conventional
interviewing techniques. As a response to
methodological and time limitations, we
constructed with settler participation a
historical baseline of pilot interventions that
have been eective in raising the bar on
individual land user economic performance
vis-à-vis the conservation of remaining
forests (Vivan et al., 2013).
Not long ago, land reform settlers taking
possession of land in the Amazon were
urged by the government to deforest half
of their lots if they wished to remain on
the land. It would thus be surprising to
nd forests conserved at the lot level, yet
we found that participating settlers have
kept on average 13 per cent more native
forest area than non-participants, even
if their average farm areas were smaller.
At the landscape level, settlements were
dierentiated in terms of remaining forests,
with a 39 per cent proportionally greater
area protected in the settlement
in Juruena versus the lowest rate
(the settlement in Juína).
The best performances are associated,
among other synergic vectors, with the
fact that the forest reserve had been
established as a collective property rather
than divided into individual lots. Settlers
in Juruena had created a cooperative
to manage these forest remnants for
Brazil nuts, the basis for a local industry.
This suggests that settlers respond to
legal restrictions when there is full set of
economic incentives present. In the other
settlements, the lack of these elements
led farmers to capitulate to dominant
economic forces in the region promoting
land-use change, practically nullifying the
demonstration eect of the ICDPs. The
existence of contradictory and constantly
shifting macro policies exacerbated weak
ties among local actors, leading settlers
to adhere to the dominant land-use trend
in the Amazon and to identify themselves
with ‘ranching culture’.
A cost–benet analysis
for sustainable cattle raising
To establish an economic rationale for
forest retention among settlers and
ranchers, research was conducted on the
extent to which forest proximity enhances
biological control of the pasture spittlebug
(Homoptera: Cercopidae), a scourge of the
cattle industry in the Amazon. Here, we
are concerned with valuing this ecosystem
service as a means to convince land users
to abide by the strictures of the forest
code. The probable values involved are
signicant: with high spittlebug infestations
(over 50 nymphs per square metre), pasture
mortality is high, and there is a pronounced
drop in stocking rates to as low as 25 per
cent of the uninfested level (Gallo et al.,
2002). Ranchers with high infestation rates
have simply abandoned their enterprises
or may be induced to deforest in other
areas in hopes of avoiding this cost when
establishing pastures.
Our method combined economic and
biological research on the relationship
between forest proximity and spittlebug
infestation rates, under dierent forest
congurations (corridors, large patches,
small remnants etc.) in Cotriguaçú. Data
on spittlebug infestation were gathered
during the outset of the rainy season in
pastures in random sampling blocks on
ve properties exhibiting dierent forest
congurations, and the distance to the
nearest forest remnant measured using
high-resolution satellite imagery.
Our results suggest that pest control
benets from forest retention in all the
studied mosaics are superior to the
We question how
best to dene an adequate
policymix to achieve both
environmental and social
targets—that is, how to halt
the degradation of natural
resources, ensuring their
sustainable use, while also
reducing poverty in the
communities where these
natural resources are located.
When addressing an issue
from a policymix perspective,
all levels of policy (national,
state and local) should be
taken into consideration.
While command and control
policies are established at a
national level, they interact
with other policies mostly at
the regional and local levels.
The International Policy Centre for Inclusive Growth
Policy in Focus 15
opportunity costs of avoiding deforestation
in these areas. Net losses from spittlebug
infestation were found to average USD120
per hectare per year. The corresponding
biological control services provided by
nearby forests were estimated to range
between USD65 and USD117 per hectare
per year, in some cases nearly reducing
pasture losses completely (Del Arco et al.,
2013). These benets were considerably
greater than the opportunity costs of
converting the forest to additional pasture.
Additional surveys among a wider group
of ranchers in Cotriguaçú pointed to
factors which have undermined eorts
to avoid deforestation. Cattle ranchers
were loath to accept payments as a
means to this end but were susceptible to
arguments regarding the agroecosystem
services provided by forest retention. Such
arguments were reinforced by pointing
to experience of local producers who
had low infestation rates associated with
forests. There is now considerable interest
in extending the research to include other
properties and a larger sample.
Conclusion: institutional
opportunities and constraints
for economic instruments
In northwest Mato Grosso, institutional
capacities to implement a complex set
of intertwined policy instruments at a
municipal level have generally been called
into question by those interviewed for
our case study. While the framework of
federal and state policies directed toward
controlling the rate and direction of land-
use change should conceivably guide
local decision-making, local governments
are rarely able to sort out contradictory
policies, public infrastructure investments
and commodity price signals and take the
lead in regularising private land use.
Although the Brazilian federal structure
permits municipalities to assume
responsibility for environmental licensing
and regulation, they are more likely to
oer concessions to large-scale enterprises
such as slaughterhouses than to impose
environmental constraints. Local political
interests are more closely aligned with
short-term resource extraction than with
sustainable enterprise aiming to benet
settlers and forest peoples.
Given these constraints, it is remarkable
that, in the course of our studies, a
number of innovative strategies to
eectively apply elements of the existing
policymix were documented such as:
the determination of a progressive
mayor in Juína who ensured that
Indigenous peoples would receive a
share of ICMS-E proceeds to promote
territorial integrity;
the time and resource commitment
of an informal army of agroforesters
and carbon sinkers (over 1000
families in northwest Mato Grosso
now have agroforestry plots on their
properties); and
the non-timber enterprise that
orchestrates minimum pricing, school
feeding procurement, value-added
tax exemptions and the voluntary
carbon market to foster resilience and
dissuade settlers from deforesting.
As a main conclusion, the cornerstone of
an eective policymix relies on a locally
constituted municipal environmental
council and fund with the capacity
required to implement the set of
instruments embodied in the new
constitutional Forest Code. Furthermore,
the ability to assist landowners with
environmental decits to identify
and trade surplus forest reserves, and
vice-versa, could be facilitated by the
development of local GIS monitoring
facilities and information management
capacities. It is of primary importance that
these developments are combined with
tenure registration in the newly created
Environmental Rural Records.
Del Arco, P., P.H. May, E. Florence and V.F.
Silgueiro (2013). ‘The eect of forest proximity
on biological control of pasture in Northwest
Mato Grosso, Brazil: a cost-benet analysis for
land use policy’, presented at the Conference of
the European Society for Ecological Economics,
Lille, France, 19 June 2013.
Ferraro, P. and A. Kiss (2002). ‘Direct payments for
biodiversity conservation’, Science, 298: 1718–9.
Gallo, D., O. Nakano, S. Silveira Neto et al. (2002).
Entomologia agrícola. Piracicaba, FEALQ.
INPE (2012). ‘Projeto PRODES. Monitoramento
da Floresta Amazônica Brasileira por Satélite’,
Instituto Nacional de Pesquisas Espaciais
website, <
php> (accessed 28 April 2014).
May, P.H., M.F. Gebara, J. Andrade, L.J. Vivan
and K. Kaechele (2012). ‘Assessment of the
role of economic and regulatory instruments
in the conservation policymix for the Brazilian
Amazon – a coarse grain study. Policymix
project report’, POLICYMIX – Assessing the role
of economic instruments in policy mixes for
biodiversity conservation and ecosystem services
provision, No. 5/2012, <http://policymix.nina.
FINAL.pdf> (accessed 28 April 2014).
May, P.H., M.F. Gebara, G. Lima et al. (2013).
‘The eectiveness and fairness of the
“Ecological ICMS” as a scal transfer for
biodiversity conservation. A tale of two
municipalities in Mato Grosso, Brazil’,
presented at the Conference of the European
Society for Ecological Economics, Lille,
France, 19 June 2013.
Minang, P.A. and M. van Noordwijk (2013).
‘Design challenges for achieving reduced
emissions from deforestation and forest
degradation through conservation: Leveraging
multiple paradigms at the tropical forest
margins’, Land Use Policy, 31: 61–70.
Ostrom, E. (1990). Governing the commons;
the evolution of institutions for collective action.
Cambridge, Cambridge University Press.
Ring, I., C. Schröter-Schlaack, D.N. Barton, R.
Santos and P.H. May (2011). ‘Recommendations
for assessing instruments in policy mixes
for biodiversity and ecosystem governance’,
Policymix Technical Brief, No. 5, <http://policymix.
(accessed 28 April 2014).
Vivan, J.L., P.C. Nunes, R. Abad, et al. (2013).
‘Pilot projects and agroenvironmental measures
in northwest Mato Grosso, Brazil: impacts and
lessons for REDD+ policy “mixes”’, presented
at the Conference of the European Society for
Ecological Economics, Lille, France, 19 June 2013.
The authors would like to acknowledge funding
from the Policymix project <http://policymix.> funded by the European Commission,
Directorate General for Research, as well as the
National Institute for Science and Technology
on Public Policy, Strategies and Development
(INCT-PPED). We would also like to express our
thanks to Paulo Nunes and UNDP/GEF project
1. Federal Rural University of
Rio de Janeiro, Brazil.
2. Federal University of Santa Catarina and
Federal Rural University of Rio de Janeiro,
Brazil (in memoriam).
3. Instituto Centro de Vida (ICV),
Mato Grosso, Brazil.
4. Federal Rural University of Rio de Janeiro
and Getúlio Vargas Foundation (Rio), Brazil.
5. Institute of Economics, Federal University
of Rio de Janeiro, Brazil.
6. ICMS-E is based on a reallocation of revenues
to municipalities from the ICMS, which is the
Brazilian sales tax charged on the transaction
of consumer goods and services.
Developing Forest-sector and REDD+
Governance: a Multi-stage, Multi-level
and Multi-stakeholder Approach in Nepal
by Federico López-Casero,1 Timothy Cadman2 and Tek Maraseni3
This article focuses on the governance
of forest carbon emissions projects and
policies. It explores how the development
of standards through multi-stage, multi-
level and multi-stakeholder processes can
contribute to ensuring good governance.
It argues that a governance standard, which
is developed through a multi-stakeholder
process at dierent levels (local, national
and international) and in several stages,
provides legitimacy to forest carbon
emissions trading. It illustrates this approach
by presenting the development of a draft
voluntary national quality-of-governance
standard for Reducing Emissions from
Deforestation and Forest Degradation and
conservation, sustainable management of
forests and enhancement of forest carbon
stocks in developing countries (REDD+)
through action research in Nepal.
Governance can be dened as the
“dynamic interplay between civil society,
business and the public sector” (Ruggie,
2003). Governance needs to address an
increasing complexity arising from its
multi-actor, multi-level (local, national, and
international) and multi-meaning nature:
dierent stakeholders may have dierent
values, interests and views (van Bodegom
et al., 2008). Therefore, multi-stakeholder
processes and social learning are required
for governance to eectively steer and
improve societal situations.
The importance of governance
in the forest sector and REDD+
Weak governance in the forest sector is
one key underlying factor or driver of
deforestation. Low levels of participation,
transparency and accountability
increase the risk of corruption, nancial
mismanagement and capture of benets
by elites—resulting in conict over forest
resources and illegal logging (WRI, 2009;
Menzies, 2007). These governance failings
result in government revenue losses of
an estimated USD10–15 billion per year
globally (ITTO, 2010).
Ensuring good governance is particularly
important in the development of a global
nancial mechanism for REDD+. Most
countries are of the view that carbon
markets will make an important contribution
to REDD+ by delivering performance-
based payments to forest owners and
managers in developing countries who
protect and/or enhance forest carbon
stocks. Millions of people live in and next to
forests worldwide, and their involvement in
REDD+ development, implementation and
governance is key to its success.
The United Nations Framework Convention
on Climate Change (UNFCCC) has
recognised the importance of good
forest governance for REDD+. In 2010
in the Cancún Agreements (Decision
1/CP.16) the 16th Conference of Parties
(COP 16) adopted social and environmental
safeguards on REDD+ that “should be
promoted and supported”—including
“transparent and eective national forest
governance structures” (COP 16 UNFCCC,
2011). However, the development,
operationalisation and institutionalisation
of forest governance denitions are largely
country-driven in response to specic
country conditions, priorities, requirements
and opportunities. Internationally
consistent governance standards will assist
governments in ensuring eective forest
institutions at the national level.
REDD+ governance quality
There have been two global-level policy
responses to governance problems facing
REDD+. One was the agreement on the
social and environmental ‘safeguards’
at COP 16 in Cancún, which has led to
requirements for stakeholder consultation
(FCPF and UN-REDD, 2012). Another
response has been the rise of ‘participatory
governance assessments’ (PGAs), which aim
at undertaking consultations to identify the
costs and benets of REDD+ to stakeholders,
and to develop safeguards (UN-REDD, 2011).
A number of social and environmental
standards for REDD+ are under
development. These include initiatives
Source: Cadman, 2011.
The International Policy Centre for Inclusive Growth
Policy in Focus 17
The need for a
comprehensive analytical
framework or standard to
assess, monitor and report
on forest governance
in REDD+ countries is
increasingly recognised at
the international level.
facilitated by the Forest Carbon
Partnership Facility (FCPF), the World Bank,
UN-REDD, and the Climate, Community &
Biodiversity Alliance (CCBA) jointly with
CARE International.
The eort that these initiatives have put
into the development of criteria to ensure
certain elements of good governance
should be acknowledged, but these are
counter-balanced by the extent to which
countries are committed to consultation
and/or have the capacity to do so. In
Panama, for example, indigenous people
recently withdrew from the national
REDD+ programme because full and
eective (i.e. meaningful) participation did
not take place (Lang, 2013).
Nepal project: objectives, research
questions and methodology
The research project in Nepal is
developing a quality-of-governance
standard to assist the eective negotiation,
development and implementation of
REDD+ but with relevance also for forest
management and emissions trading
schemes in general. The key features of
the standard development process are
that it is multi-stage, multi-level and multi-
stakeholder. The project uses a common
framework of principles, criteria and
indicators (PC&I) of governance that:
ensure consistent and comprehensive
governance in REDD+ development
and implementation; and
reect national circumstances and
stakeholder requirements (see Table 1).
The primary method for the standard
development project has to involve key
forest-sector and REDD+ stakeholders,
typically including government, forest user
groups, other civil society organisations,
minorities and international aid programmes.
The multi-stakeholder approach ensures the
representation and involvement of all key
sectors of forestry and REDD+ in developing
the standard. The stakeholders are engaged
throughout a number of stages to identify
site- and context-specic veriers of
governance quality at the local, subnational
and national levels.
The active involvement and participation
of a diverse range of stakeholders
demonstrated that many key groups
and individuals were able to experience
the value of developing such a standard
in a collaborative environment, which
fostered meaningful participation and
resulted in productive deliberation
around a whole series of core governance
challenges including inclusiveness, equality,
transparency, accountability, decision-
making and implementation. Particular
Source: Authors’ elaboration.
emphasis has been placed on facilitating
the involvement of marginalised groups
who seldom have the opportunity to
participate in such processes. The approach
creates governance standards that are
likely to have a high degree of local
ownership and relevance (see Figure 1).
The need for a comprehensive analytical
framework or standard to assess, monitor
and report on forest governance in REDD+
countries is increasingly recognised at the
international level, including the UNFCCC.
Ensuring emissions reduction through
good governance is vital for the longer-
term viability of REDD+.
Several initiatives have developed
governance standards for REDD+, but they
have not been developed through genuine
multi-stakeholder processes, in the sense of
stakeholders providing the contents of the
standards as active participants throughout
all stages of the process.
Rather than making the stakeholders the
‘subjects’ of governance, the Nepal project
has ensured that all major stakeholder
groups have had the opportunity to
identify what they felt was needed to
ensure good governance. Context-specic
standards have the advantage of making it
easier for all participants to determine what
they require in a given local, subnational
The Necessity of Land Governance:
Sustainable Development in the Amazon
by Bastiaan Philip Reydon1 and Vitor Bukvar Fernandes1
There are many dimensions to
sustainable development in the
Amazon, but prevention of primary
forest deforestation is undeniably the
most important. The Brazilian Forest Code
has demonstrated that Brazil is unable to
establish clear policies about deforestation
and that debates have been supercial and
ideological in nature. The preservation of
the Amazon rainforest requires deeper and
more signicant changes, which should
be more comprehensive than dicult-to-
enforce laws and regulations.
It is undeniable that the strong command
and control policies2 implemented in the
last few years have played a crucial role
in reducing deforestation. Due to their
nature, command and control policies
depend on direct state intervention and
can hardly be enforced for a long period
of time. This is because the main causes of
deforestation—activities such as livestock
herding, agricultural production and power
1. Natural Resources and Ecosystem Services
Area, Institute for Global Environmental
Strategies, Hayama, Japan.
2. Institute for Ethics, Governance and Law,
Grith University, Queensland, Australia.
3. Australian Centre for Sustainable
Catchments, University of Southern
Queensland, Queensland, Australia.
or national situation before policies and
projects are developed. The governance
framework and involvement methodology
used, however, can also be applied for the
development of governance standards
elsewhere in the world.
Cadman, T. (2011). ‘Quality and legitimacy
of global governance. Case lessons from
forestry’, International Political Economy Series.
Basingstoke, Palgrave Macmillan.
Conference of the Parties 16 (COP16), United
Nations Framework Convention on Climate
Change (UNFCCC). (2011). Decisions adopted
by the Conference of the Parties 1/CP.16: The
Cancun Agreements: Outcome of the work
of the Ad Hoc Working Group on Long-term
Cooperative Action under the Convention,
Report of the Conference of the Parties on
its sixteenth session, held in Cancun from 29
November to 10 December 2010. <http://unfccc.
pdf#page=2> (accessed 7 April 2014).
FCPF and UN-REDD (2012). Guidelines on
Stakeholder Engagement in REDD+ Readiness
State in Community-Based Forest Management.
New York, Columbia University Press.
Ruggie, J.G. (2003). ‘Taking Embedded
Liberalism Global: The Corporate Connection,
in D. Held and M. Koenig-Archibugi (eds),
Taming Globalisation: Frontiers of Governance.
Cambridge, Polity Press: 93.
UN-REDD (2011). ‘REDD+ Participatory
Governance Assessments Piloted in Indonesia
and Nigeria’, Programme Newsletter, No. 20:
July, <
(accessed 24 April 2014).
with a Focus on the Participation of Indigenous
Peoples and Other Forest-Dependent Communities.
Washington, DC, Forest Carbon Partnership
Facility and Geneva, UN-REDD Programme,
PDF/July2012/Guidelines on Stakeholder
Engagement April 20, 2012 (revision of March
25th version) (1).pdf> (accessed 24 April 2014).
ITTO (2010). ‘Getting a lock on governance’,
Tropical Forest Update, Vol. 19, No. 1. Yokohama,
International Tropical Timber Organization,
(accessed 24 April 2014).
Lang, C. (2013). ‘COONAPIP, Panama’s
Indigenous Peoples Coordinating Body,
Withdraws from UN-REDD’, redd-monitor.
org website, 6 March 2013, <http://www.
Monitor%29> (accessed 24 April 2014).
Menzies, N. (2007). Our Forest, Your Ecosystem,
Their Timber. Communities, Conservation, and the
generation—will continue to exist; therefore,
permanent solutions must be adopted.
A greater use of economic incentive
policies3 to avoid deforestation, such as
those proposed in the Forest Code, is
one of the alternatives that has been
discussed the most, both in literature and
in social movements.4 The most important
economic instrument in this regard is
payment for environmental services (PES).
Studies indicate that the main opportunity
cost to be compensated by PES relates to
the productive gains associated with the
land.5 Andrade (2007), using a literature
review as a basis, and Fasiaben (2008),
relying on studies conducted in Acre,
came to the conclusion that the
approximate average amount paid
per year as compensation to avoid
deforestation was USD100 per hectare.
Wunder et al. (2009) propose an opportunity
cost based on the alternative use of the
forest (traditionally timber, livestock and
grain) converted into equivalent carbon
dioxide, using the carbon market as a
basis, and reaching values up to BRL671
(approximately USD288)6 per hectare.
Nevertheless, both Fasiaben (2008) and
Wunder (2008) highlight the importance
of suitably regulating/controlling land
ownership in order to use PES to
preserve the forest.
As demonstrated by Reydon (2007),
one of the most important incentives
for deforestation is the increase in land
value as a result of forest clearing.
Data from AgraFNP7 initially reveal, as
shown in Table 1, that the prices of land
with forest coverage vary from state
to state, costing from BRL108 per
hectare in the state of Acre to BRL546
in Mato Grosso.
It is also noteworthy that in less
deforested states (Acre, Amapá and
The International Policy Centre for Inclusive Growth
Policy in Focus 19
Amazonas) the price of land is lower
than in Mato Grosso, Para and Rondonia,
which have higher rates of deforestation.
Despite this, the most signicant
conclusion drawn from Table 1 is that
deforestation always signicantly
increases the value of land. Moreover, on
average, prices are four times higher in
cleared areas. This occurs because land
pricing is essentially a consequence of
the expectations of obtaining productive
gains from farming activities undertaken
in those areas. This rationale is based on
the fact that deforested areas may be
used immediately, with no need to incur
additional clearance expenses.
In the most extreme case (Acre), these
prices are 14 times higher for deforested
areas, and in the state of Amazonas
they are almost 10 times higher. Few
investments yield results as high as these.
Reydon and Fernandes (2011) used a case
study containing primary data from the
municipality of Cotriguaçú to demonstrate
the signicant increase in land prices as
a result of deforestation. Amounts varied
from BRL878.12, using primary data, to
over BRL1000, according to secondary data.
Thus the laws that prohibit deforestation
(the Forest Code and the Environmental
Crimes Act) are ineective, given that
deforestation still occurs and generates
extraordinary prots from the appreciation
of land. This happens because of impunity,
particularly in this area.
Consequently, it may be concluded that
the most important mechanism to avoid
deforestation is the elimination of land
speculation, both in deforested areas and
areas undergoing forest clearance.
This means that the State must enforce
clear land policies to control land use and
prevent the predatory and speculative
use of these areas, as demonstrated by
Deininger (2003), Reydon (2007) and
FAO (2009), among others. In the case
of Brazil, Reydon (2011) also shows that
the integration and improvement of the
dierent property registries maintained
by a variety of institutions is the rst
step that must be taken to implement a
comprehensive, integrated land policy.
The existence of a single land governance
system capable of integrating all
competent entities—the registry oces
(public notaries), the National Institute of
Colonization and Land Reform, the state
land institutes and other agencies such
as the Department of Federal Revenue—
may trigger a process to control land
property and use in the country, including
mechanisms to monitor deforestation and
illegal public land appropriation, especially
in the Amazon rainforest.
Therefore, apart from focusing on PES
to reduce deforestation, we must also
forestall the arbitrary use of land and
prevent owners from clearing these
areas to obtain illegal prots.
In summary, land regulation and
governance are the rst few measures to be
adopted when implementing any type of
policy meant to preserve forests. Therefore,
the way forward is to create clear and
Source: AgraFNP (2009).
objective land regulations and governance
mechanisms to reduce the speculative use
of land. Only then can solutions to enable
PES be eectively structured.
Andrade, J.P.S. (2007). ‘A implantação do
pagamento por serviços ecossistêmicos no
Território Portal da Amazônia: uma análise
econômico-ecológica’, Masters dissertation.
Campinas, Instituto de Economia, Universidade
Estadual de Campinas.
AgraFNP (2010). Anualpec. Relatório de Análise do
Mercado de Terras. Sao Paulo, AgraFNP.
Deininger, K. (2003). Land Policies for Growth
and Poverty Reduction. Washington, DC,
World Bank and Oxford University Press.
FAO Land Tenure and Managment Unit
(2007). ‘Buena gobernanza en la tenencia y
la administración de tierras’, Estudios sobre
Tenencia de La Tierra, No. 9. Rome, Food and
Agricultural Organization of the UN.
Fasiaben, M.C.R., D.C. Andrade, B.P. Reydon,
J.R. Garcia and A.R. Romeiro (2009). ‘Estimativa
de aporte de recursos para um sistema de
pagamento por serviços ambientais na
oresta amazônica brasileira’, Ambiente e
Sociedade, Vol. 12: 0622.
Margulis, S. (2003). Causas do Desmatamento
da Amazônia Brasileira. Brasília, World Bank.
Reydon, B.P. (2007). ‘A regulação institucional
da propriedade da terra no Brasil: uma
necessidade urgente’ in P. Ramos (ed.), Dimensões
do Agronegócio Brasileiro: Políticas, Instituiçõe e
Perspectivas. Brasília, MDA (NEAD – Estudos 15).
Reydon, B.P. (2011). ‘La questión agraria brasileña
necesita gobernanza de tierras’, Land Tenure Journal
Revue des Questions Foncières, Vol. 1: 127–147.
Reydon, B.P. and V.B. Fernandes (2011).
Mercados de Terras de Cotriguaçú: subsídios
para o cálculo do custo de oportunidade da oresta
preservada. Relatório de Pesquisa. Cuiabá, ICV.
Wunder, S. (2008). ‘Payments for environmental
services and the poor: concepts and preliminary
evidence’, Environmental and Development
Economics, 13: 279–297.
always signicantly
increases the value of
land. Moreover,
on average, prices are
four times higher in
cleared areas.
Integrating Livelihoods and Land-use
Change at the Frontiers of Deforestation
by Aldicir Scariot1
The conversion of areas of native
vegetation into agricultural land is the main
driver of immediate change in land-use,
which in turn negatively aects biodiversity,
ecosystem services and water resources and
contributes to climate change.
Environmental degradation aects the
natural environment’s capacity to recover
and to sustain future production processes,
which ultimately depend on functioning
ecosystems. Smallholder farmers, indigenous
peoples and traditional populations, who
depend signicantly on natural goods and
services for their livelihoods and income,
are the most aected. In this context of
rapid and intense loss of biodiversity
and ecosystem services, the debate over
strategies that reconcile food production,
biodiversity conservation and ecosystem
services has focused on two alternatives
treated as incompatible.
Proponents of a land sharing strategy
argue that biodiversity conservation and
food production can be integrated in the
same area, using more environmentally
friendly farming methods. The contrasting
alternative would be land sparing, with the
separation of the areas for production from
the areas for conservation. In land sparing,
agriculture would be intensied, targeting
high yields and increased productivity.
In this strategy, the intensication of
agriculture would use smaller areas for
production, thus saving areas to protect
the remaining natural habitats from
agricultural expansion.
This debate has focused mainly on
biodiversity conservation versus food
production, but has neglected the
existence of indigenous peoples,
traditional extractivists and family farmers.
Indigenous peoples occupy 22 per cent of
the land surface of the world and 11 per
cent of forested areas (Sobrevila, 2008).
Smallholder farmers and traditional
and indigenous peoples supply their
subsistence needs and generate income
from family farming and harvesting of
forest products, hunting and shing.
These peoples depend on functioning
ecosystems and natural areas with which
they interact and which they aect.
Management proposals which state
that saving land is more suitable for the
tropics, where vast areas of intact forest
are threatened by agricultural expansion,
ignore the number of people who live
in such forests and those who, although
they do not live within, depend on
forests for their survival. About 1.6 billion
people in rural areas depend on forests to
dierent degrees (FPP, 2012). According
to the World Commission on Forests and
Sustainable Development, 350 million of
the world’s poorest people depend almost
entirely on forests for their subsistence and
survival. A further 1 billion poor people
depend on remnant woodlands, trees
in homestead gardens and agroforestry
systems for fuel wood, food and fodder
needs. Worldwide 60 million indigenous
peoples and other communities live
and depend on forests for subsistence
(Krishnaswamy and Hanson, 1999).
In Brazil, the largest tropical country,
where indigenous people make up a small
proportion of the population, indigenous
lands—which collectively cover 12.5
per cent (106.7 million hectares) of the
country’s surface area—are home to more
than 300 indigenous groups speaking
more than 250 dierent languages.
In these regions, most of the indigenous
land habitats are preserved, generally
characterised by marked biodiversity,
and part of the area cultivated with crops,
managed for hunting and the extraction
of timber and non-timber forest products.
Governments are beginning to recognise
that dierent people interact in dierent
ways with the environment, and are
implementing actions to ensure that
these dierences are recognised in public
policy through access to and use of land
and resources. In Brazil, public policies
recognise the existence of more than a
1 Institute of Economics, University of
Campinas, Brazil.
2. The main command and control policies, direct
state interventions that modify the behaviour
of those who deforest, were: a) Operations
Curupira (2005) and Arco de Fogo (2008), which
fought illegal logging; b) Decree 6321/07, which
restricts bank credit and forces land owners in
the municipalities with the highest levels of
deforestation to re-register; c) the creation of
Conservation Units (UCs) adding over 20 million ha
to the 80 million ha already in existence, amounting
to a total of 273 UCs; d) the ratication of 87
Indigenous Territories and approximately 18 million
ha; and e) restrictions on produce originating
from properties located in municipalities with the
highest levels of deforestation.
3. The economic incentive policies, which are
initiatives that use economic mechanisms
(pricing or others) to encourage economic
agents to reduce deforestation or to prevent
their action, included Operation Arco Verde
(2008), a special credit line from the regional
development funds for the Northeast (FNE), the
North (FNO) and the Central-West (FCO) meant
to promote the recovery of degraded areas,
reforestation, management and environmental
regularisation in the Legal Amazon.
4. See Andrade (2007) Fasiaben (2008),
Wunder (2008) and others for a good
review of the literature.
5. Margulis (2002 and 2003) shows that livestock
activities generate signicant earnings in the
Amazon and, as such, are some of the major
causes of deforestation. This nding opposes
most of the literature on the topic, which
indicates that livestock gains were associated
only with subsidies.
6. Exchange rate as of 25 March 2014:
USD1 = BRL2.32847.
7. A leading business information specialist
focusing on agriculture and food policy.
The International Policy Centre for Inclusive Growth
Policy in Focus 21
Photo: Laercio Miranda. Village of Barranco Vermelho, Erikbaktsa Indigenous Land; Sor ting of Brazil nuts,
storage shed in the background.
About 1.6 billion
people in rural areas
depend on forests.
Different people interact
in different ways with the
environment, and are
implementing actions
to ensure that these
differences are recognised
in public policy.
dozen traditional peoples, with special
arrangements for the granting and tenure
of land. Traditional peoples are dened
according to the way they occupy a
territory, use natural resources as part of
their identity and whether the exploitation
of natural resources is sustainable and
adapted to local ecological conditions.
More than 17 per cent of Brazil’s territory
is dedicated to the conservation
of biodiversity through the federal
conservation system. Of this area 51.6
million hectares, 6.1 per cent, are spared
for integral protection, where commercial
use or natural resource extraction
activities are prohibited.
The sustainable use category covers
11.1 per cent (94.4 million hectares) of
the country, and denotes areas where
traditional people can hunt, sh, harvest
non-timber forest products and cultivate
crops, conserving biodiversity and
ecosystem services. In addition, 5.2 million
private farms in the country, covering
330.2 million hectares should, by law,
allocate 20–80 per cent of their area
for sustainable use and conservation,
according to the ecosystem in which they
are located, with intensive agriculture
allowed on the remaining area.
This mixture of shared land-use,
represented by the sustainable use and
conservation of land and strictly protected
areas, is the foundation of land-use
in Brazil. Given its large availability of
arable land and the potential for yield
improvements, Brazil can contribute a
large fraction of the increase in production
to meet the world demand for food until
2050 (Lapola et al., 2014). Furthermore,
Brazil has the opportunity to promote
a new land-use paradigm for tropical
countries that are heavily dependent on
agriculture, safeguarding a considerable
fraction of the world’s biodiversity in intact
forests and savannahs while guaranteeing
signicant food production to meet
world demand. It is, therefore, crucial
that land-use change be guided by
solid sustainability principles.
Nowhere is this challenge more daunting
than at the development frontiers, such as
the northwest of the state of Mato Grosso
in Brazil, where rapid land-use changes
are associated with intense environmental
degradation and social conicts. But how
could such future land-use be sustainably
conducted if this challenge is not
confronted where it is most intense?
Protecting areas and sharing the land
are equally important and should be
used as complementary strategies at the
landscape level. Recognising the complex
relationship that rural communities have
with the natural areas in the tropics
is critical to succeed in planning and
implementing public policies that aim
to conserve biodiversity and ecosystem
services and respect the livelihoods of
indigenous, traditional communities
and family farmers. The conservation
of biodiversity in tropical agricultural
landscapes requires a combination of
land sharing and land sparing, and should
be shaped by the relevant socio-cultural
conditions. It is, therefore, dependent
on the context. This mixed strategy is
particularly important in the frontier
areas, where there are opportunities for
intelligent management of the landscape
that integrates the various economic and
conservation activities while respecting
local ways of life.
Fischer J., P. Batáry, K.S. Bawa et al. (2011).
‘Conservation: limits of land sparing’,
Science, 334: 593–94.
FPP (2012). Forest peoples: numbers
across the world. Moreton-in-Marsh, UK,
Forest Peoples Programme.
Krishnaswamy A. and A. Hanson (1999).
Our forests, our future: summary report.
Winnipeg, World Commission on
Forests and Sustainable Development,
(accessed 7 April 2014).
Lapola, D.M., L.A. Martinelli, C.A. Peres,
J.P. Ometto, M.E. Ferreira, C.A. Nobre and
I.C. Vieira (2014). ‘Pervasive transition of the
Brazilian land-use system’, Nature Climate
Change, 4(1): 27–35.
Sobrevila, C. (2008). The role of indigenous
peoples in biodiversity conservation: the natural
but often forgotten partners. Washington,
DC, World Bank.
1. Embrapa, Genetic Resources and
Biotechnology, Brasília, Brazil.
Beyond the Panacea: a Critical Assessment
of Instruments of Deforestation Control
by Raoni Rajão,1 Britaldo Soares-Filho,2 Camilla Marcolino,1 Richard van der Ho 1 and Marcelo Costa1
Since the late 1980s the Brazilian
government has adopted various
instruments aimed at controlling
deforestation in the Amazon. Despite
marked dierences from a conceptual
and practical perspective, when these
instruments were introduced, the
government often considered them to be
eective and inherently superior solutions
to those previously used to control
deforestation. In this article we shall assess
these instruments critically, to emphasise
their results, potential and limitations.
This analysis is based on 84 interviews
and extensive direct observations of the
procedures conducted by dierent Brazilian
government entities between June 2007
and July 2009. In addition to qualitative
data, this study also drew on quantitative
geographical data provided by INPE (the
National Institute for Space Research), IBAMA
(the Brazilian Institute of Environment and
Renewable Natural Resources) and SEMA-
MT (the Secretary of the Environment of the
State of Mato Grosso).
The following three sections will present
and evaluate institutional, economic as
well as command and control instruments,
respectively. The nal section will
highlight the need to use these tools in an
integrated fashion, to unite their strengths
and mitigate the weaknesses of each type
of deforestation control modality.
Command and control instruments
Command and control instruments
are undoubtedly the most widely used
environmental governance modality in
the Amazon. In addition to serving as
guiding principles for IBAMA’s activities
since the institution was created, the
instruments were also adopted by various
environmental agencies at the state level
(Órgãos Estaduais do Meio Ambiente—
OEMAs). A mechanism is considered a
command and control mechanism when
the government rst ‘commands’ the
development of a given environmental
law and then ‘controls’ its enforcement
through inspection activities (Stewart,
1996). Although widely used for controlling
deforestation in the Amazon since the
Programa Nossa Natureza (‘Our Nature
Programme’) in 1988, it was not until the
2000s that these instruments reached
the scale needed to produce an eect on
reducing deforestation (Mello, 2006).
The hiring of additional IBAMA personnel
in the form of environmental analysts with
university degrees—and the creation of
DETER (a real-time deforestation detection
system), by INPE, played a key role on this
front (Rajão and Vurdubakis, 2013).
According to INPE data, the period
between 2004 and 2012 registered
an 83 per cent decrease in Amazon
deforestation. Although this decline
occurred concomitantly with two editions
of the Plan for the Prevention and Control
of Deforestation in the Legal Amazon area
(PPCDAm), it has not yet been possible to
determine the precise role played by that
policy in the decline of deforestation. In
any case, the preliminary results of our
study of the impact of IBAMA’s actions
on the decline in deforestation showed a
statistically signicant correlation between
the number of nes levied by IBAMA in
a given municipality and the decline in
deforestation in that same location in the
following year.
Despite these positive results, it is
important to highlight the nancial and
regulatory limitations of command and
control instruments in the region. From a
nancial standpoint, IBAMA’s enforcement
actions carry a high operational cost
and rely on the skills of inspectors to
be eective and successful (Rajão and
Vurdubakis, 2013). For example, we may
consider the total number of infraction
notices issued by IBAMA across the country
between 2004 and 2007 and the budget
spent by the institute to develop a rough
indicator. This amounts to BRL204,805
(about USD90,000) per infraction notice
(IBAMA, 2008). Although this amount is just
an approximation and does not include
other activities carried out by IBAMA
(e.g. licensing, environmental education
etc.), the magnitude of this gure suggests
that command and control actions do
indeed entail high transaction costs.
Consequently, any attempt to expand
these activities will meet budget and
infrastructure constraints and will be
hampered by the inability of inspection
bodies to cover the entire territory.
For example, the sum of all areas ned
by IBAMA for illegal deforestation amounts
to only 17.21 per cent of the total area
deforested between 2004 and 2008.
Given that only a small fraction of
oenders eventually suer sanctions,
inspection activities are seen by ned
farmers as arbitrary and unfair. This
means that the command and control
instruments come into conict with the
notions of justice and fairness considered
central to enabling modern forms of
governance (Foucault, 1977;
Weber, 1922/1968).
Furthermore, since the amount of people
ned is very low, the deterrence eect
this action should have does not manifest
itself satisfactorily. Thus, it is unfeasible to
promote a legitimate and stable order in
Amazon land management based solely
on command and control.
Institutional instruments
Institutional instruments constitute
another branch of environmental
governance mechanisms widely discussed
in the context of the Amazon. This type
of instrument has a broad denition that
often overlaps with command and control
as well as economic mechanisms. Despite
this challenge, we can characterise
institutional mechanisms as a typology
of environmental governance conducted
through a set of rules and political and
administrative structures that indirectly
contribute to policy objectives.
As such, while command and
control mechanisms directly enforce
The International Policy Centre for Inclusive Growth
Policy in Focus 23
environmental laws and punish those who
disobey them a posteriori, institutional
mechanisms try to oer a legal and
administrative context to encourage
compliance with the law, to avoid nes
before they are imposed. Examples of such
mechanisms in the context of the Amazon
include land regulation programmes
(e.g. Terra Legal, Decree 6992/2009),
socio-economic environmental zoning
(Ab’Saber, 1989) and the establishment
of special protection areas, such as
indigenous, extractive and environmental
conservation reserves.
The creation of protected areas was
undoubtedly one of the most eective
measures to control deforestation in
the last decade (Nepstad et al., 2006).
Regarded as one of the main pillars of the
PPCDAm, a sizeable number of protected
areas were created by the government
between 2004 and 2009, covering 54 per
cent of remaining Amazon rainforests.
Consequently, Soares-Filho et al. (2010)
estimate that the creation of protected
areas accounted for 37 per cent of the
reduction in deforestation witnessed
between 2004 and 2006.
Environmental licensing and registration
are other types of institutional instruments
used widely in the Amazon in recent
years. These tools are based on geo-
referencing and the use of satellite imagery
to determine the environmental status
of rural holdings. Using these records,
government control entities are expected
to carry out farm inspections using satellite
images and to hold oenders accountable
for environmental crimes.
Conceptually, environmental licensing and
registration instruments can be considered
ideal forms of social control, due to
their potential to carry out inspections
that are both universal (i.e. everyone in
the system may be subjected to it) and
have low transaction costs (i.e. the use
of satellite imagery abolishes the need
for on-site visits when issuing notices
for illegal deforestation activities). As
such, these systems could, theoretically,
provide a foundation for the development
of a disciplinary type of environmental
governance—they are, therefore, seen by
the population as modern and legitimate
systems (Foucault, 1977).
However, an analysis of the eectiveness
of the licensing system for rural properties
(SLAPR) in the state of Mato Grosso has
shown that these objectives are not always
achieved. In particular, a comparison
between deforestation inside and outside
the system suggests that this instrument
has contributed to increasing—rather than
decreasing—deforestation within
licensed properties.
One of the reasons for this may be the
registration strategy, which leaves it up
to owners to choose which properties
will be included in the system, leading
to the exclusion of properties with
environmental liabilities and the inclusion
Photo: UNDP Brazil. Aripuana River, Northwestern Mato Grosso.
The creation
of protected areas was
undoubtedly one of the
most effective measures to
control deforestation
in the last decade.
of properties with vegetation coverage,
for purposes of obtaining deforestation
approval. Moreover, the state agency
has not systematically used the ability
to remotely control deforestation
throughout the duration of the study.
(Rajão, Azevedo and Stabile, 2012).
Economic instruments
Finally, the third type of mechanism
is characterised by encouraging
environmentally sustainable behaviours by
providing positive incentives, usually of a
nancial nature (Juras and de Araújo, 2008).
Some of the policies that use economic
mechanisms are: the ecological ICMS,
which transfers funds to municipalities
according to ecological indicators (Ring,
2008); the clean development mechanisms
(CDMs), created by the Quito Protocol of
1997 (Austin et al., 1999); carbon credit for
reducing emissions from deforestation and
forest degradation—commonly known as
the UN-REDD;3 and incentive programmes
for sustainable production (Le Tourneau
and Greissing, 2010; Lederer, 2011).
These mechanisms are based on the notion
that economic players who deliberately
decide to reduce their environmental
impact should be nancially compensated,
directly and indirectly (Fearnside, 1997;
Kaimowitz, 2008; Olsen, 2007).
Such cases can already be found in existing
literature, where economic incentives to
sustainable development have become
viable and long-term alternatives for
local populations (Le Tourneau and
Greissing, 2010). For example, the Project
for Conservation and Sustainable Use of
Forests in North-western Mato Grosso
(the UNDP/GEF project) showcases the
multiplying nature of these initiatives,
which bring together an increasing
number of local stakeholders and
disseminate sustainable economic
practices related to the extraction
of latex and Brazil nuts (Tito, Nunes
and Vivan, 2011).4 However, two important
limitations of these initiatives are their
relatively small scale and the pressure
put on these areas by domestic and
international markets to increase the
production of, mostly, beef and soybeans—
which, historically, have been linked to
deforestation (Hargrave and Kis-Katos,
2013). In this context, REDD was seen by
several stakeholders as a way to obtain
enough funding to oset these economic
pressures and encourage the preservation
of the forests (Kaimowitz, 2008;
Nepstad et al., 2009.).
However, expectations of receiving large
amounts of funds through REDD have
not yet materialised. Several factors could
explain the diculties faced by REDD,
the most apparent of which is the lack of
consensus within the several Conferences
of the Parties (COP) of the United Nations
Framework Convention on Climate Change
(UNFCCC) about the mechanism and the
nancial crisis in Europe and the
United States.
There are also internal inconsistencies
within the mechanism—while some
players see REDD as an economically
ecient market instrument, others see
it as a government policy detached
from the market. Furthermore, there are
profound dierences in issues related to
the methodology used in monitoring and
calculating credits and in fund allocation,
and well as in the protection of biodiversity
and indigenous rights (IPAM, 2011).
Even within existing REDD projects related
to the voluntary carbon market there are
problems in the criteria used to calculate
the credits, since the future baseline
methodology used by such projects tends
to project future deforestation levels
well above historical gures, to achieve
a higher level of ‘avoided’ deforestation
and, consequently, increase prots from
the sale of credits (Lang, 2013; Leach and
Scoones, 2013). In addition, current models
do not consider the fact that deforestation
trajectories can change dramatically
according to national and international
contexts (Soares-Filho, Lima, Bowman
and Viana, 2012).
These methodological controversies are
compounded by the emergence of local
movements against REDD, reecting the
existence of deeper concerns about the
negative consequences of this mechanism,
such as increased social inequality, for
example (Arsel and Büscher, 2012).
Final thoughts
In short, all of the deforestation control
instruments currently enacted—to a
greater or lesser extent—by the Brazilian
government have their limitations.
In particular, we have seen that when
command and control instruments are
scaled up, they quickly run into logistical,
nancial and legitimacy limitations.
Similarly, institutional instruments cannot,
by themselves, bring about behavioural
changes—and can even be used for
adverse purposes.
Despite being greatly emphasised in recent
years, economic instruments have had
trouble expanding their local sustainable
development activities to regional levels and
securing substantial nancial resources—
from REDD or other mechanisms.
On the other hand, each of these
instruments also has its strengths, and
they have been key in bringing about
positive advances. For example, IBAMA’s
inspection activities have had a signicant
eect on the decline of deforestation.
Likewise, the creation of protected areas
has contributed to curbing deforestation
by creating obstacles to the possession
of public lands without a clearly dened
purpose. Moreover, local projects to
promote the production of latex and Brazil
nuts, even on a small scale, have made it
possible to combine a higher income and
better quality of life within environmental
conservation activities.
As such, by combining dierent
approaches to controlling deforestation
one can arrive at a “policy mix” where the
weaknesses of the dierent instruments
can be mitigated, thus building synergies.
Particularly, a reduction in the opportunity
cost of environmental preservation can
also be observed in areas with eective
command and control structures.
Similarly, even though the licensing of
rural properties has not had the expected
eect, these records facilitate inspection
work and lower the transaction
costs of enforcement.
Finally, the existence of economically
viable alternatives to deforestation—along
with improved command and control
policies—changes the risk–reward
relationship of environmental crimes
and contributes to the establishment
of sustainable practices. This analysis
suggests that the search for an “optimal”
solution to deforestation is an ambition
doomed to fail. Therefore, one must
adopt a comprehensive strategy that
takes into account the need to integrate a
heterogeneous set of public policies.
The International Policy Centre for Inclusive Growth
Policy in Focus 25
1. Laboratory of Environmental Services
Management, Federal University of Minas Gerais.
2. Centre of Remote Sensing, Federal University
of Minas Gerais.
3. The United Nations Reducing Emissions from
Deforestation and Forest Degradation (REDD)
is an initiative to produce a nancial value for
the carbon that is stored in forests by providing
incentives for countries to reduce emissions
from forested areas and to invest in low-carbon
pathways of development.
4. Also see Vivan et al. in this issue.
Ab’saber, A. (1989). ‘Zoneamento ecológico e
econômico da Amazônia: questões de escala e
método’, Estudos Avançados, 3(5): 4–20.
Arsel, M. and B. Büscher (2012). ‘Nature TM
Inc.: changes and continuities in neoliberal
conservation and market-based environmental
policy’, Development and Change, 43(1): 53–78.
Austin, D., P. Faeth, R.S. Da Motta, C. Ferraz, C.E.F.
Young, Z. Ji et al. (1999). ‘How much sustainable
development can we expect from the clean
development mechanism?’, Climate Notes.
Washington, DC, World Resources Institute: 1–16.
Fearnside, P.M. (1997). ‘Environmental services as
a strategy for sustainable development in rural
Amazonia’, Ecological Economics, 20(1): 53–70.
Foucault, M. (1977). Discipline and punish: the
birth of the prison. Penguin Social Sciences.
Hargrave, J. and K. Kis-Katos (2013). ‘Economic
Causes of Deforestation in the Brazilian
Amazon: A Panel Data Analysis for the 2000s’,
Environmental and Resource Economics,
54(4): 471–494.
IBAMA (2008). ‘IBAMA em números’, IBAMA
website, , (accessed 2 April 2014).
IPAM (2008). REDD in Brazil: a focus
on the Amazon. Brasília, Centro de
Gestão e Estudos Estratégicos.
Juras, I. and S. de Araújo (2008). ‘Instrumentos
econômicos de política ambiental e reforma
tributária’, Cadernos ASLEGIS, 33(1): 109–127.
Kaimowitz, D. (2008). ‘The prospects for Reduced
Emissions from Deforestation and Degradation
(REDD) in Mesoamerica’, International Forestry
Review, 10(3): 485–495.
Lang, C. (2013). ‘Disney’s commitment to Mickey
Mouse REDD: Conservation International’s
trick baseline for the Alto Mayo project in Peru’,
REDD-monitor website, <http://www.redd-
project-in-peru/> (accessed 2 April 2014).
Le Tourneau, F.-M. and A. Greissing (2010).
‘A quest for sustainability: Brazil nut gatherers
of Sao Francisco do Iratapuru and the
Natura Corporation’, The Geographical
Journal, 176(4): 334–349.
Leach, M. and I. Scoones (2013). ‘Carbon forestry
in West Africa: The politics of models, measures
and verication processes’, Global Environmental
Change, 23 (5) 957–967.
Lederer, M. (2011). ‘From CDM to REDD+ —
What do we know for setting up eective and
legitimate carbon governance?’, Ecological
Economics, 70(11): 1900–1907.
Mello, N. (2006). Políticas territoriais na Amazônia.
Sao Paulo, Annablume.
Nepstad, D., S. Schwartzman, B. Bamberger,
M. Santilli, D. Ray, P. Schlesinger et al. (2006).
‘Inhibition of Amazon deforestation and re
by parks and indigenous lands’, Conser vation
Biology, 20 (1): 65–73.
Nepstad, D., B.S. Soares-Filho, F. Merry, A. Lima,
P. Moutinho, J. Carter et al. (2009). ‘The end of
deforestation in the Brazilian Amazon’, Science,
326(5958): 1350–1351.
Olsen, K.H. (2007). ‘The clean development
mechanism’s contribution to sustainable
development: a review of the literature’, Climatic
Change, 84(1): 59–73.
Rajão, R., A. Azevedo and M.C.C. Stabile (2012).
‘Institutional subversion and deforestation:
learning lessons from the system for the
environmental licensing of rural properties
in Mato Grosso’, Public Administration and
Development, 32(1): 229–244.
Rajão, R. and T. Vurdubakis (2013). ‘On
the pragmatics of inscription: detecting
deforestation in the Brazilian Amazon’, Theory,
Culture & Society, 30(4): 151–177.
Photo: Laercio Miranda. Village of Barranco Vermelho, Erikbaktsa Indigenous Lands; Brazil Nuts Integrated
Programme (PIC). Experience interchange meeting.
The existence
of economically
viable alternatives to
with an improved ability
to command and
control—changes the
risk–reward relationship
of environmental crimes
and contributes to
the establishment of
sustainable practices.
Ring, I. (2008). ‘Integrating local ecological
services into intergovernmental scal transfers:
the case of the ecological ICMS in Brazil’, Land
Use Policy, 25(4): 485–497.
Soares-Filho, B.S., P. Moutinho, D. Nepstad, A.
Anderson, H. Rodrigues, R.A. Garcia et al. (2010).
‘Role of Brazilian Amazon protected areas in climate
change mitigation’, Proceedings of the National
Academy of Sciences, 107(24): 10821–10826.
Soares-Filho, B.S., L. Lima, M. Bowman and
L. Viana (2012). Challenges for Low-Carbon
Agriculture and Forest Conservation in Brazil.
Washington, DC, Inter-American
Development Bank.
Stewart, R.B. (1996). ‘The future of environmental
regulation: United States environmental
regulation: a failing paradigm’, Journal of Law
and Commerce, 15: 585–596.
Tito, M.R., P.C. Nunes and J.L. Vivan (2011).
Desenvolvimento agroorestal no noroeste
do Mato Grosso: dez anos contribuindo para a
conservação e uso das orestas. Brasília, UNDP.
Weber, M. (1922/1968). Economy and society:
an outline of interpretive sociology. New York,
Bedminster Press.
Deforestation in the Himalayas:
Myths and Reality
by Jean-Marie Baland1 and Dilip Mookherjee2
Deforestation in developing and
middle-income countries is an urgent
global problem, aecting climate
change, soil erosion, changes to major
river basins and the livelihoods of poor
households living near forests. Public
discussions of the problem are frequently
dominated by widely held beliefs
concerning the extent of deforestation
(that it is large and growing over time)
and its impacts on local livelihoods
(that these are adverse and also
signicant). Views concerning
determinants of deforestation include
economic growth, local poverty and
inequality, all of which are generally
believed to accelerate the process.
Of possible remedies, the one most
widely discussed involves property rights
over forests: that local communities
should be granted ownership and
management autonomy in order to
adequately arrest deforestation.
There are many good reasons as to
why these propositions could be true,
informed both by economic theory and
casual empiricism. Human populations
use forests for household energy, as an
excess food resource for livestock, and
timber for wood products. Forest areas
are often cleared to extend agricultural
cultivation, increase mining exploration,
create residential construction projects
or expand urban and peri-urban areas.
Economic growth that increases demand
for food, energy, mineral resources,
furniture and housing could thus
naturally increase deforestation. Among
those living near forests, the poorest
households rely the most on forests
for firewood, fodder as well as other
forest products and resources. They rely
more on livestock grazing, are less able
to afford commercial fuels or timber
and have numerous family members
(especially women and children) with
a low opportunity cost of time who
can collect forest products. Hence
increased poverty among neighbouring
populations could increase human
pressure on the forests.3
Increased deforestation could, therefore,
have a severe impact on local poverty,
possibly generating a vicious cycle
wherein this increased poverty may in
turn accelerate deforestation. Women
and children, the principal collectors, are
likely to be the most adversely aected.
Greater socio-economic inequality of
local communities could undermine their
capacity to engage in collective action to
impose and enforce limitations on forest
use. Shifting ownership rights over forests
to local communities away from the State
might, therefore, enhance the scope and
power of such collective action.
These views are commonly expressed
in numerous anecdotes, media reports,
academic research, and policy documents
of national governments and international
organisations. However, to what extent are
they upheld by results of empirical ground-
level research? Do they apply equally to
dierent countries or continents?
In collaboration with various
researchers over the past decade,
we have undertaken a study of the mid-
Himalayan region spanning Nepal and
northern India, using a variety of detailed
micro-level data sets. For Nepal we have
relied on three successive rounds of the
nationally representative household
Living Standards Measurement Survey
(LSMS) between 1995 and 2010. For
the Indian states of Himachal Pradesh
and Uttaranchal, which fall in the same
geo-climatic zone as Nepal, we carried
out detailed household, community and
forest surveys between 2001 and 2004.
The ndings turn out to be similar across
Nepal and the two Indian states, as well
as with studies in these regions by a
number of other researchers.
Increasing deforestation?
There is no clear evidence that
deforestation in this part of the world is
accelerating. For India as a whole, Foster
and Rosenzweig (2003) use aerial satellite
data on forest biomass and nd the
opposite phenomenon of reforestation.
Our detailed ground-level forest surveys
in Himachal and Uttaranchal indicate that
the key problem is degradation rather than
deforestation. Tree branches are heavily
lopped, stunting tree growth and limiting
foliage. Some 61 per cent of forest areas
sampled exhibited canopy cover below the
ecologically sustainable threshold of 40 per
cent. In contrast, measures of tree biomass
were not alarmingly low: mean basal area
exceeded the sustainability threshold of
40 m2 per hectare. While forest areas have
receded owing to growing encroachments,
this accounts for a relatively small fraction
of the increased time taken by households
to collect rewood. Over the past quarter
of a century, rewood collection times
have increased by 60 per cent on average,
but walking time to the forest increased by
only 10 per cent. The bulk of the increased
collection time was due to the declining
quality of the forest, with households
taking longer to nd rewood because
trees were being more heavily lopped.
These facts imply that a feasible research
strategy for testing various hypotheses
concerning determinants of forest
degradation is to study their eects on
household rewood and fodder use, and on
the quality of neighbouring forests. In this
regard, the main ndings are the following:
Eects of economic growth
Here what we nd depends on the
precise way that ‘growth’ is measured.
If it is measured in terms of household
consumption levels, the evidence (based
on estimated household Engel curves)
shows that economic growth aggravates
degradation: rising consumption levels
(up to the 95th percentile) are associated
with increased rewood collection/
use. However, the same is not true
when growth is measured in terms
of key household productive assets
rather than consumption levels. Only
The International Policy Centre for Inclusive Growth
Policy in Focus 27
growth in livestock assets seems to
have a strong positive impact on the
demand for rewood. The eect of land
ownership is negligible, and education
and non-farm assets have a negative
eect. Indeed, in villages of Nepal, per
household collections of rewood fell
between 1995 and 2010, explained
mainly by rising education and non-farm
assets, shrinking livestock numbers and
greater outmigration. Hence the nature
of growth matters. If it is accompanied
by occupational changes, with local
populations shifting from traditional
livestock or land-based occupations
to modern non-farm occupations, a
reduction in forest degradation can occur.
The opposite may happen if growth
in living standards is driven by rising
income transfers from the government or
remittances, or by rising livestock assets.
Far more important than economic
growth in explaining trends in forest
degradation in the Himalayan region
were demographic factors, such as
the rise in population and increasing
fragmentation of rural households.
Shrinking household size, a growing
population and slow rates of permanent
out-migration have translated into fast
growth in the number of rural households,
increasing forest degradation. A 10 per
cent growth in productive assets in the
two northern Indian states was estimated
to raise household rewood use by
less than 0.2 per cent, while a 10 per
cent growth in population was
estimated to raise it by 9.9 per cent.
Eects of local poverty
There is no evidence that poor households
collect more rewood than non-poor
households. In reality it is the other way
around. Non-poor households have
greater energy needs, related to house size,
consumption of cooked foods, and of heat
during the winter. This result is robust with
respect to estimation methodology and
applies to both Nepal and northern India.
Declining poverty is, therefore, unlikely to
arrest forest degradation.
Eects of forest
degradation on local poverty
The data also show very limited evidence
for the reverse link between forest
degradation and current living standards
of forest neighbouring populations. An
increase in rewood collection time by
an hour in northern India (comparable
to the extent observed over the past
quarter of a century) was estimated to
lower household consumption by less than
1 per cent uniformly across poor and non-
poor households. The reason is that the
opportunity cost of households doing
so time-wise is low, since they accumulate
rewood during lean agricultural seasons.
It is possible, however, that there will
be some adverse eects on local
livelihoods in the long run, if current
degradation trends continue.
Eects of local
inequality or collective action
Neither is there any evidence that
increased inequality of consumption or
land ownership in forest neighbouring
villages is associated with greater pressure
on adjoining forests. Informal collective
action to regulate forest use in northern
India is conspicuous by its absence, except
in a few locations. This does not reect a
general inability to engage in collective
action, as indicated by functioning informal
cooperatives in the context of other local
public goods, such as irrigation or temples.
Part of the reason could be the fact cited
above: a more degraded forest has a
negligible impact on current household
livelihoods. So local communities do not
worry about the condition of neighbouring
forests nor do they try to regulate
the use of forest products.
Eects of local community
ownership and management
Both Nepal and India have transferred
ownership and responsibility for
the management of forests to local
communities, in the form of forest-
user-groups (FUGs) in Nepal and Van
Panchayats (VPs) in northern India.
These local organisations have created
and enforced rules for rewood and
fodder use by their members, and
engaged in reforestation programmes.
While estimating the impact of
these changes raises a number of
methodological problems, the most
detailed studies available nd a 10–20
per cent reduction in household rewood
use in either region. In northern India
these ndings are corroborated by
reduced lopping of forests transferred to
VPs, compared with neighbouring state
and open-access forests.
Eects of varying costs
of household energy substitutes
Our studies in northern India show that
household use of rewood is sensitive
to the cost of alternate modern fuels,
especially liquid petroleum gas (LPG).
A subsidy of INR100 (approximately USD2,
one third of the cost in the early 2000s)
of an LPG cylinder was estimated to
reduce household rewood use by
around 20–27 per cent.
In summary, many commonly held
views—such as the eects of economic
growth, poverty reduction or local
inequality on forests, or of the reverse
eects of forest degradation on local
livelihoods—turn out either to be invalid
or require serious further qualication in
the Himalayan context.
What appears to be true instead is the
following. Forest degradation is a serious
problem, from the standpoint of its
Photo: Sandesh Timilsina. Collecting Forest Fodder. Rokuwa, Nawalparasi, Nepal.
Property Rights, Deforestation and Violence:
Problems for the Development of the Amazon
by André Albuquerque Sant’Anna1 and Carlos Eduardo Frickmann Young2
In Brazil the loss of forested areas has
historically been connected to the process
of land occupation and to the production
methods established at the beginning of
the colonial era (Young, 2006). From this
historical perspective, despite the change
in the type of goods that generated the
dynamics of the colonial economy and,
subsequently, the imperial and republican
economies, one notices a ‘rise and fall’
movement resulting from the direct or
indirect exploitation of natural resources.
The abundance of a specic resource
causes its rapid predatory exploitation,
which consequently leads to a long-term
decline caused either by the growing
scarcity of what was once plentiful, or by
an abrupt increase in the supply of a given
product, prompting a trend of continuous
price depreciation in domestic and
international markets.
Based on this analytical perspective, a
theory concerning the economic cycles
1. Centre for Research in the Economics of
Development, University of Namur, Belgium.
2. Department of Economics, Boston University,
United States.
3. Editor’s note: One should keep in mind that
this is a salient issue in the forest frontiers of Asia
discussed in this article, which have a very high
population density. Therefore, some of the issues
discussed here should not be automatically
applied to all forest frontiers, especially in
the Amazon frontier case, which has a very
low population density and is dominated by
agricultural commodities.
larger, non-local ecological and climate
change impacts, as well as possible
long-term impacts on local livelihoods.
It results mainly from firewood and
fodder use by households that live near
the forests. Informal collective action
by neighbouring local communities is
unlikely to offer a meaningful solution
for the problem. Transfer of ownership
and management to local communities
is, however, likely to help moderate
firewood use and encourage forest
regeneration. Subsidies and increasing
availability of modern energy substitutes
will reduce household reliance
on the forest.
In the long run, the most eective means
of limiting degradation will be policies that
control population growth and promote
education, growth of non-farm occupations
and permanent out-migration.
Baland, J.M., P. Bardhan, S. Das and D.
Mookherjee (2010). ‘Forests to the People:
Decentralization and Forest Degradation in
the Indian Himalayas’, World Development,
38 (11): 1642–1656.
Baland, J.M., S. Das and D. Mookherjee
(2011). ‘Forest Degradation in the Himalayas:
Determinants and Policy Options’, in S. Barrett,
K.G. Maler and E. Maskin (eds), Environment
and Development Economics: Essays in Honour
of Sir Partha Dasgupta. Oxford, Oxford
University Press.
Baland, J.M., P. Bardhan, S. Das, D. Mookherjee
and R. Sarkar (2008). ‘Environmental Impact of
Poverty: Evidence from Firewood Collection
in Rural Nepal’, Economic Development and
Cultural Change, 59 (1) 2010: 23–61.
Baland, J.M., P. Bardhan, S. Das, D. Mookherjee
and R. Sarkar (2007). ‘Managing the
Environmental Consequences of Growth:
Deforestation in the Himalayas’, in S. Bery, B.
Bosworth and A. Panagariya (eds), India Policy
Forum 2007. London, Sage Publications.
Baland, J.M., F. Libois and D. Mookherjee
(2012). ‘Firewood Collections and Economic
Growth in Rural Nepal 1995–2010: Evidence
from a Household Panel’, Working Paper,
No. 247. Boston, MA, Institute for Economic
Development, Boston University.
Foster, A. and M. Rosenzweig (2003). ‘Economic
Growth and the Rise of Forests’, Quarterly Journal
of Economics, 118: 601–637.
Somanathan, E., R. Prabhakar and B.S. Mehta
(2009). ‘Decentralization for cost-eective
conservation’, Proceedings of the National
Academy of Sciences of the United States of
America, 106 (11): 4143–4147.
covering the period between the 15th
century and the beginning of the 20th
century (pau-brasil, sugar, cattle, gold
and coee) can be devised, despite the
enormous dierences in the modes of
production and distribution for these
products. All these activities were
associated with the predatory use of
natural resources, causing severe damage
to the environment without, however,
creating sustainable methods to overcome
the economic and social contradictions.3
Nevertheless, conservative groups with
considerable inuence in the Brazilian
National Congress and other entities
still advocate that the expansion of the
agricultural frontier is paramount for
the development of remote areas of the
country. According to this mindset, the
transformation of the forest into pastures
or plantations would enable economic
gains because of the increase in farming
activities. Moreover, the process of
territorial occupation would intensify
the presence of the State in areas of the
country that still lack public goods.
The argument stating that the geographical
expansion of agriculture leads to
development is neither new nor exclusive
to Brazil. In the USA, for instance, there are
those who maintain that the expansion
of the frontier in the march towards the
West was crucial not only for the economic
growth of the country but also for the
promotion of a democratic society.
In fact, according to Turner, “American
democracy is fundamentally the outcomes
of the experiences of the American people
in dealing with the West” (1920).
From this standpoint, called the ‘Turner
Thesis’ or ‘Frontier Thesis’, the success or
failure of the frontier expansion policy
depends on the level of institutional
development of the country. The empirical
exercise by García-Jimeno and Robinson
(2009) endorses the hypothesis, referring
The International Policy Centre for Inclusive Growth
Policy in Focus 29
Land property
rights in Brazil—and
especially in the
Amazon region—
are structured in
a perverse manner.
Social instability
and the absence of
suitable infrastructure for
basic services, such
as education and
health, are also
consequences of an
unbalanced expansion
caused by the ‘production
of property rights’
through deforestation.
to the ‘conditional frontier’. On the
one hand, in countries where political
institutions were weak during the period
of frontier expansion, there is a negative
relationship between the frontier and
current per capita income.
On the other hand, in countries with more
advanced political institutions—such as
the USA and Canada, for instance—the
result is precisely the opposite: there is
a positive association between frontier
expansion and per capita income.
Nevertheless, these countries still maintain
a high percentage of forest coverage
(54 per cent coverage in Canada and 33
per cent in the USA), indicating that if, in
fact, deforestation played a role in these
countries’ development, this was only true
until the middle of the 20th century.
In light of this scenario, it is quite important
to understand the role of institutions,
especially with regards to property rights, in
the advance of deforestation in Brazil and
its consequences for development. Our
position stems from the assumption that
the inadequate denition of land property
rights is a key factor for deforestation and its
perverse social and economic eects.
In frontier areas there is no formal
denition of land ownership. Additionally,
Brazil follows a tradition in which land
ownership can only be claimed based
on the productive use of the land. In this
context, land deforestation is almost
considered mandatory to increase the
chances of obtaining land ownership.
Be that as it may, once property
values rise, land grabbers usually
claim ownership rights.
Usually the dispute between land-
appropriators and land-grabbers leads
to intense conicts: literature reveals
that in municipalities where levels of
deforestation are high, there is also more
violence, as measured by the homicide
rate (Sant’Anna and Young, 2010).
Therefore, it may be concluded that
ongoing violence is the clearest result of
an inadequate denition of land property
rights in the expansion of the Brazilian
agricultural frontier.
In a scenario of intense land concentration,
the expansion of the agricultural
frontier towards forest areas has been
used historically as a safety valve to
accommodate the surplus population
that resulted from the mechanisation and
demobilisation of family farming (Sant’Anna,
2012). Consequently, initially the Atlantic
rainforest and subsequently the Cerrado
region and the Amazon rainforest were
reduced to accommodate land conicts.
However, no land reform was implemented
in previously occupied areas. Nonetheless,
newly occupied areas reproduce the same
land concentration pattern. Thus, after
some time, the inability to absorb this new
surplus recreates the conditions for a new
migration ow towards unconverted forest
areas (Young, 2006).
Since the way in which the land economy
is organised in Brazil poses limits to
the capacity of obtaining ‘natural’ land
surpluses to accommodate social
problems, the use of frontier lands as an
element to mitigate this issue is far from
a denitive solution. Therefore, this is an
expansion model that inevitably
leads to deforestation, violence and
land concentration.
On the other hand, the expansion of
pasture lands and plantation areas has a
devastating impact on native forests, while
failing to produce a socially desirable
situation: most of the poverty-stricken
areas of the country are located in rural
regions where deforestation is
already consolidated.
Empirical studies demonstrate that
deforestation is not associated with an
increase in the Human Development Index
(HDI): Young and Neves (2009) show that
in the municipalities in which the Atlantic
rainforest was more severely deforested
in the period between 1985 and 1996 the
HDI growth was lower than in most other
cities. Celetano et al. (2009) also reveal
that there is no relationship between
the percentage of deforested areas in a
municipality in the Amazon region
and any increase in its HDI.
Social instability and the absence of
suitable infrastructure for basic services,
such as education and health, are also
consequences of an unbalanced expansion
caused by the ‘production of property
rights’ through deforestation.
Health problems are quite severe, given
that the loss and degradation of the native
vegetation increases the risk of the spread
of disease (UNDP, 2010). Literature shows
that deforestation contributes to the spread
of infectious diseases such as malaria,
dengue fever, Chagas disease, leishmaniasis
and the hantavirus. This situation is made
worse by climate change. Consequently,
there is a direct rise in public expenditures
because the costs of mitigation and
eradication strategies are higher than those
of preventive actions. It is estimated that
expenses to ght malaria in the Americas
amounted to over USD500 million between
2004 and 2007 (UNDP, 2010).
It becomes clear in this context that land
property rights in Brazil—and especially
in the Amazon region—are structured
in a perverse manner. The process of
land concentration, associated with the
mechanisation of agriculture, forces
a considerable part of the population
into moving to the agricultural frontier.
The confusing institutional framework
leads to the process of ‘production of
property rights’, which entails continuous
deforestation and violence associated
with land conicts. Thus, public policies
must be improved so as to reduce
incentives for deforestation and make
better use of consolidated lands.
Celentano, D., E. Sills, M. Sales and A. Veríssimo
(2009). ‘Deforestation and human development:
evidence of boom-bust development in the
Social Policies and Forest Frontiers:
the Consequences for Agricultural
Land-use in the Brazilian Amazon
by Gabriel Henrique Lui1
Rural populations around the world
are currently experiencing a process
of livelihood transition. Especially for
smallholder farmers, this process is
characterised by diversication and a
disconnection between rural livelihoods
and agricultural yields. In general, this
scenario is a result of factors such as:
(1) unequal competition with large-scale
farmers; (2) low protability of subsistence
farming; (3) new job opportunities and
o-farm activities; (4) proximity to cities
and urban culture; (5) a lack of interest
among younger generations to continue
agricultural activities; (6) environmental
degradation; and (7) the lack of available
land (Bryceson, 1996; Graziano da Silva,
1997; Rigg et al., 2001; Rigg, 2006).
As a consequence, income from agricultural
activities has been shown to be decreasing
in importance among smallholder farmers,
whereas income from other activities, such
as part-time jobs, services, government jobs,
family remittances and other benets, are
becoming more important. In Latin America,
for example, Reardon et al. (2001) indicated
that up to 40 per cent of rural households’
income comes from non-agricultural
activities. In South Asia, Gordon (1999)
found that this gure can reach
upwards of 60 per cent.
In Brazil there are two signicant variables
in this scenario of livelihood diversication
among low-income smallholder farmers:
the Programa Bolsa Familia (PBF) and rural
retirement schemes. The PBF is a federal
government programme that undertakes
conditional cash transfers to 13 million
families—about 53 million people or 27
per cent of the Brazilian population. The
main goal of the PBF is to reduce poverty
and ensure food security among the
programme beneciaries. The PBF targets
families who have a monthly per capita
income of up to BRL140 (about USD62),
and conditionality is related to both health
care and school attendance. The amount
received by each family can vary between
BRL32 and BRL306 per month, depending
on the number and age of children (Brazil,
2012a). Besides the large proportion of the
population served by the programme, the
impact of additional resources becomes
clearer when considering changes in
income. Families in the programme have
experienced an average increase of 48.7
per cent in per capita income (BRL48.69
to BRL72.42), and the benet represents
about 33 per cent of total household
income on average (ibid.).
Another factor that is bringing about
signicant changes to incomes in rural
areas of Brazil is the broader access to
social security benets, such as retirement
plans, pensions and temporary assistance
(maternity, death, unemployment etc.).
The number of such beneciaries in the
country grew from 22 million in 2003
to 28 million in 2010 (Brazil, 2012b).
In rural areas, specically, the number
of beneciaries grew from 7 million to
8.4 million throughout the same period.
More than the number of individuals
served by social security programmes,
it is interesting to note the amount of
resources available from such benets.
The minimum wage, which is the basis
for calculating payments, increased from
Brazilian Amazon’, presented at the IV Congress
of the Asociación Latinoamericana y del Caribe de
Economistas Ambientales y de Recursos Naturales.
Heredia, Universidad Nacional de Costa Rica.
Dean, W. (2004). A ferro e fogo: a história e a
devastação da Mata Atlântica brasileira, 1st edition.
Sao Paulo, Cia. das Letras (1st print 1996).
García-Jimeno, C. and J.A. Robinson (2009).
‘The Myth of the Frontier’, NBER Working Paper,
No. 14774. Cambridge, MA, National Bureau
of Economic Research.
Sant’Anna, A.A. (2012). ‘Land inequality
and deforestation in the Brazilian Amazon’,
Annals of the 40th ANPEC Annual Meeting.
Porto de Galinhas, Associação Nacional dos
Centros de Pós-Graduação em Economia,
ecc9245102.pdf> (accessed 25 March 2014).
Sant’Anna, A.A. and C.E.F. Young (2010).
‘Property Rights, Deforestation and Rural
Conicts in the Amazon’, Economia Aplicada,
Vol. 14, No. 3: 377–387.
Turner, F.J. (1920). The frontier in American history.
New York, H. Holt and Co.
UNDP (2010). ‘América Latina e o Caribe:
Uma Superpotência de Biodiversidade’,
(Portuguese); ‘Latin America and the
Caribbean: A Biodiversity Superpower’,
Policy Brief. New York, UNDP,
Policy_Brief_ENG.pdf> (English)
(accessed 25 March 2014).
Young, C.E.F. (2006). ‘Desmatamento e
Desemprego Rural na Mata Atlântica’,
Floresta e Ambiente, 13(2): 75–88.
1. BNDES - The Brazilian Development Bank.
The author is exclusively responsible for the
opinions expressed in this article.
2. Institute of Economics, Federal University of
Rio de Janeiro, Brazil.
3. See Dean (2004) for a complete description
of the process of occupation of the Brazilian
territory and its socio-environmental impacts.
Young, C.E.F. and A.C.M. Neves (2009).
‘Destroying the Myth: Deforestation, Rural
Employment and Human Development in the
Brazilian Atlantic Forest’, presented at the IV
Congress of the Asociación Latinoamericana
y del Caribe de Economistas Ambientales y
de Recursos Naturales. Heredia, Universidade
Nacional de Costa Rica.
The International Policy Centre for Inclusive Growth
Policy in Focus 31
Source: Author’s elaboration. Note: * Composition of household income in 2003 and 2011, showing an inversion
in the proportion of retirements/pensions and agriculture in the period.
The effects of
the PBF and rural
retirement emerged,
therefore, as one of
the components that
contribute to the broader
phenomenon of a
reduction in small-scale
agricultural activities in the
Amazon region.
BRL240 per month in 2003 to BRL545 per
month in 2011, representing a real growth
of 49 per cent. The higher number of
beneciaries, coupled with the rise in the
value of benets, increased the economic
importance of these income sources. For
example, in the region of Santarém, in
the eastern Amazon, the annual amount
of resources transferred from retirement
benets and pensions grew from BRL96
million in 2003 to BRL242 million in 2010
(ibid.), which represents more than 13 per
cent of the GDP in this area.
Faced with new income sources such as
the PBF and rural retirement benets that
do not require the mobilisation of labour,
what type of decision-making logic would
households follow? Would there be a
movement to reinvest additional income
in agricultural activities? Or would these
external income sources contribute to the
further detachment of these families from
agriculture? Consequently, how do such
decisions aect land-use?
To address such questions, two primary
data surveys were conducted by
researchers from Indiana University
(USA) and the University of Sao Paulo
(Brazil) in the region of Santarém,
in the state of Para. The surveys were
undertaken in 2003 (n = 488) and 2011
(n = 83), seeking demographic, economic
and spatial information for the same
properties on both occasions. Data were
collected through questionnaires and
semi-structured interviews. Land-use
change was analysed through satellite
imagery for 1997, 2001, 2005 and 2010.
Quantitative data were analysed using
descriptive statistics, correlation analysis
and hypothesis testing. Furthermore,
qualitative data were analysed by
categorising the content of interviewees’
responses (D’Aantona et. al, 2006;
Lui, 2013; Lui and Molina, 2013).
Results show signicant changes between
the two periods, such as a reduction in
average household size (from 4.6 to 3.86
people), a growth in monthly income
(from BRL443.97 to BRL1463.17) and
expenditures (BRL344.12 to BRL933.60),
a decrease in the production of annual
crops, especially rice (-82 per cent), beans
(-73 per cent) and corn (-63 per cent), and
an acceleration in the rate of deforestation
on large properties, especially after the
arrival of large-scale soybean producers.
The growing importance of non-
agricultural income sources for households,
especially from retirement benets and
pensions, was one of the most prominent
results of this work. Besides being an
essential source for family subsistence,
as evidenced by income data (Figure 1),
the movement of resources generated by
retirees and pensioners has fundamental
consequences for the local economy
and represents a signicant part of
municipal GDP.
The continuous increase in the minimum
wage and the value of retirement
benets tends to further reduce the
share of agriculture in the composition
of household income. The same logic
applies to o-farm jobs, with the increase
in the minimum wage making them more
attractive than the income generated by
agricultural activities.
The eects of the PBF and rural retirement
emerged, therefore, as one of the
components that contribute to the broader
phenomenon of a reduction in subsistence
agricultural activities in the Amazon region.
However, this phenomenon is also strongly
inuenced by internal dynamics, such as
the reduced availability of family labour,
the depreciation of agricultural work, and
the search for educational opportunities
elsewhere as well as external jobs. It is
also aected by external dynamics, such
as low cash returns for the main annual
crops produced by subsistence farming,
competition with large, mechanised farms,
Sustainable Settlements in the Amazon
by Mauro Angelo Soave Junior,1 Osvaldo Stella Martins,1 Paulo Roberto de Souza Moutinho1 and Simone Mazer Rodrigues1
The challenge of transitioning family
farming into a low carbon economy
Greenhouse gas emissions caused by
land-use change are responsible for about
12 per cent of global emissions or 1.2
billion tons per year (Le Quéré et al., 2009).
In Brazil, 55 per cent of greenhouse gas
emissions are linked to land-use change,
and take place mainly in the Amazon
(MCTI, 2013). These emissions are mostly
linked to the process of conversion of
forest areas into croplands.
Although deforestation in the Amazon
has decreased dramatically in the last
decade, obvious diculties remain in
preserving the rainforest. In this sense,
the Amazon poses a great challenge:
to build a rural development model
that is aligned with the prospect
of zero deforestation and improved
quality of life for farmers.
One of the key elements in this process
is that of agrarian reform settlements.
Deforestation in agrarian reform
settlements in the Amazon
Currently, about 78 per cent of settlement
areas in Brazil are located in the Amazon.
These settlements cover an area of 35.7
million hectares and have the capacity
to accommodate approximately 400,000
families (IPAM, 2012). The establishment
of settlements in the Amazon region has
been growing signicantly since 2005.
In 2010, in the western region of the
state of Para, there were 216 recognised
the increased occurrence of agricultural
pests, transportation costs of production,
and the role of middlemen in the process
of commercialisation.
Most families interviewed did not observe
advantages in continuing agricultural
activities. It makes more sense to choose
other activities, since the opportunity
cost of labour increases when facing
the possibilities of formal employment,
which then reduces the prospect
of choosing agriculture as a primary
source of income. For families who are still
engaging in agriculture, eective technical
assistance seems to be the dierence,
as it is able to increase the potential of
cash returns, possibly to the same level
as wages from formal employment.
The role of technical assistance was
highlighted both in the results of
statistical tests as well as by interviewees.
Despite this scenario of detachment from
agriculture, improvements in infrastructure,
especially the increased access to energy
and water, have encouraged families to
continue living in rural areas. There is also
a feeling of belonging and of appreciation
for rural life, especially among elderly
people. Most rural landowners show
no interest in moving to urban areas.
New income sources and improved
infrastructure in rural areas has led to a
disconnection between living standards
and agricultural activities.
Considering the processes in progress,
the short- and medium-term prospects
for smallholder farmers will depend
on a scenario of economic growth,
specialisation of agriculture and the
inuence of urban culture and values,
especially among younger individuals.
It will be necessary to promote eective
technical assistance, aimed at reducing
costs, increasing production eciency
and added value so that households can
remain landowners. Furthermore, it is
important to appreciate the social and
economic role of smallholder farmers in
oering food to the broader population,
and their environmental role in resisting
the complete commodication of these
rural regions, which would certainly
accelerate deforestation. These actions,
however, cannot ignore that non-
agricultural activities are a fundamental
part of the livelihoods of the rural
population in Brazil, noting that the
interest and welfare of the families
cannot be underestimated.
Brazil (2012a). ‘Bolsa Família’, Ministério de
Desenvolvimento Social e Combate à Fome
website, <>
(accessed 3 April 2014).
Brazil (2012b). ‘Anuário Estatístico da Previdência
Social – AEPS’, Ministério da Previdência Social
website, <
php?id=423> (accessed 3 April 2014).
Bryceson, D.F. (1996). ‘Deagrarianization and
Rural Employment in sub-Saharan Africa:
A Sectoral Perspective’, World Development,
Vol. 24, No. 1: 97–111.
D’Antona, Á.O., L.K. Vanwey and C.M. Hayashi
(1996). ‘Property Size and Land Cover Change
in the Brazilian Amazon’, Population and
Environment, 27: 373–396.
Gordon, A. (1999). ‘Diversity in rural incomes:
issues aecting access at household level’, World
Bank Seminar on the Rural Non-farm Economy.
London, Natural Resource Institute.
Graziano da Silva, J. (1997). ‘O novo rural
brasileiro’, Nova Economia, Vol. 7, No. 1: 43–81.
Lui, G.H. (2013). Transformação de modos
de vida rurais na Amazônia: uma perspectiva
longitudinal sobre diversicação da renda,
atividades agrícolas e uso da terra entre
pequenos produtores’. Doctoral thesis,
interuniversity post-graduate programme in
Applied Ecology, Universidade de Sao Paulo.
Lui, G.H. and S.M.G. Molina (2013). ‘Benefícios
Sociais e Transição de Modos de Vida Rurais:
uma análise do Bolsa Família e da aposentadoria
rural entre pequenos produtores na Amazônia’,
Política & Trabalho, Vol. 38: 137–155.
Reardon, T., J. Berdergue and G. Escobar
(2001). ‘Rural nonfarm employment and
incomes in Latin America: overview
and policy implications’, World Development,
Vol. 29, No. 3: 395–409.
Rigg, J. (2006). ‘Land, farming, livelihoods, and
poverty: Rethinking the links in the Rural South’,
World Development, Vol. 34, No. 1: 180–202.
Rigg, J. and S. Nattapoolwat (2001).
Embracing the global in Thailand: activism
and pragmatism in an era of deagrarianization’,
World Development, Vol. 29, No. 6: 945–960.
1. Ministry of the Environment, Brazil.
The International Policy Centre for Inclusive Growth
Policy in Focus 33
Sustainable Settlements in the Amazon
Source: Authors’ elaboration.
settlement projects, of which 212
were federal projects and four were
state projects. These settlements are
coordinated by the oce of the Instituto
Nacional de Colonização e Reforma Agrária
(INCRA—National Institute for Colonisation
and Agrarian Reform) in Santarém, its
three advanced units (Altamira, Itaituba
and Monte Alegre) and the Instituto de
Terras do Para (ITERPA—Land Institute of
the State of Para); they have the capacity
to accommodate up to 108,300 families,
occupying a total area of 97,800 km2 of
forest and oodplain areas.
A study conducted by the Instituto de
Pesquisa Ambiental da Amazônia (IPAM—
Amazon Environmental Research Institute),
in partnership with INCRA, compared
deforestation data from PRODES2 (INPE,
2012) with data from 1868 settlements in
the Amazon biome. IPAM considered these
settlements as one large group, without
taking into account internal boundaries,
since such data were not available. By 2012,
36 per cent of the settlement area (12.8
million hectares) had been deforested.
However, most of these areas were already
deforested before being included in
agrarian reform projects. On average, the
settlements created since 1997 had already
had 38 per cent of their area deforested
(IPAM, 2012). The study also showed that
the declining deforestation trend observed
in the region as a whole is also observed in
the agrarian reform settlements (Figure 1).
The challenge of a productive transition
towards a low carbon economy
In general, settlements in the Amazon
are located in areas abundant in forest
resources, and as such, are constantly
subject to dispute over the timber and
non-timber resources they contain. For the
forestry sector, logging in the settlements
is an easy way to obtain timber legally,
considering that command and control
actions have been restricting the supply
of illegal timber to the local market, and
that forest concession policies have been
advancing very slowly in the region.
In the context of the region’s sustainable
development, a current major challenge
is to consolidate a policy of sustainable
settlements. To this end, strategies need
to be adopted to ght the drivers of
deforestation and forest degradation
by implementing a new productive
model that reduces pressure on the
remaining forest and increases
regional economic potential.
There are currently two key drivers of the
deforestation process in settlements in
the Amazon: uncontrolled logging and
extensive agricultural production.
The threat of misuse and depletion of
forest resources within settlements,
caused by uncontrolled logging (which
is illegal and has a high environmental
impact), is present either prior to the
settling of families, when the areas are
vulnerable to be looted by illegal
loggers, or even after settlement.
Due to poor infrastructure, precarious
facilities and a lack of planning within
settlements, families are left without basic
living conditions and resort to selling or
exchanging their forested lands for road
developments and other benets, or even
abandoning the area completely.
In turn, agricultural practices that
dominate the settlements are still marked
by the low productivity of the extensive
raising of livestock and a slash-and-burn
system of agricultural expansion. Yet, in
such ‘primitive’ production systems that
still prevail in much of the Amazon, the
forest plays the role of a ‘farm subsidy’,
providing nutrients and acting as a
control mechanism against crop pests,
diseases and weeds.
Therefore, encouraging forest
management and intensifying
agricultural production in areas that
have already been cleared and are
abandoned, or areas under extensive
production systems, are strategic
actions to tackle forest degradation and
deforestation in Amazon settlements.
The project ‘Sustainable settlements
in the Amazon: the challenge of
transitioning family farming into a
low carbon economy’, the result of a
partnership between IPAM, Live Produce
and Preserve Foundation (FVPP) and
INCRA, aims to promote a transition of
production systems of agrarian reform
settlements in the Amazon. The goal is to
turn the predominant production systems
currently characterized by high carbon
emissions and low protability into low
emission systems that are also protable
for local farmers. The aim is to implement
systems that maintain the carbon stocks
of forest cover within the settlements, to
increase the protability of areas already
cleared and to promote the improvement
of the environmental quality of the region.
The project, which receives nancial
support from the Amazon Fund, operates
in three agrarian reform settlement
Source: Authors’ elaboration.
projects located in western Para (PA
Bom Jardim, PA Moju and PA Cristalino II),
assisting approximately 2700 families
in an area of 230,000 hectares (Map 1).
It is scheduled to run until 2017. To meet
the project’s objectives, activities focus on
the following areas: environmental and
land regulation; strengthening of shared
management (co-management); transition
of production systems; processing and
marketing of forest products; payment
for environmental services (avoided
deforestation); sustainability indicators
and monitoring; and dissemination of
project activities (communications).
The legalisation of the environmental
situation contemplated by the land
regulation activities, coupled with the
strengthening of the capacity of settlers
to manage their own property, means
and resources, is a way to promote
their empowerment, so that when the
project comes to an end its results can
be replicated and improved over time.
This is a starting point for changing the
implementation models of agrarian reform
settlements in the Amazon and their
connection with deforestation.
Brazil (2006). Law No. 11.326 of 24 July 2006.
Brasília, Federal Government of Brazil.
Instituto de Pesquisa Ambiental da Amazônia
(2013). Amazônia em Pauta, No. 1. Brasília,
(accessed 17 April 2014).
Instituto Nacional de Colonização e Reforma
Agrária (2012). Boletim Assentamentos Verdes,
No. 1. Brasília, INCRA, <
(accessed 17 April 2014).
Instituto Nacional de Pesquisa Espacial
(2012). ‘Projeto PRODES’, INPE website,
(accessed 17 April 2014).
Le Quéré, C., M.R. Raupach, J.G. Canadell
and G. Marland (2009). ‘Trends in the sources
and sinks of carbon dioxide’, Nature Geoscience,
Vol. 2: 831–836.
Ministério da Ciência, Tecnologia e Inovação –
MCTI (2010). Segunda Comunicação Nacional
do Brasil à Convenção Quadro das Nações Unidas
sobre Mudança do Clima. Brasília, MCT.
Ministério do Desenvolvimento Agrário
(2010). O encontro da agricultura familiar
com a alimentação escolar. Cartilha, MDA.
Pacheco, P. (2009). ‘Agrarian Reform in the
Brazilian Amazon: Its Implications for Land
Distribution and Deforestation’, World
Development, Vol. 37, No. 8: 1337–1347,
Projeto de Monitoramento do
Desorestamento na Amazônia Legal (PRODES)
(2011). ‘Taxas anuais do desmatamento de
1988 até 2011’, INPE website, <http://www.>
(accessed 17 April 2014).
Tourneau, F. and M. Bursztyn (2010).
‘Assentamentos Rurais na Amazônia: Contradições
Entre a Política Agrária e a Política Ambiental’,
Ambiente e Sociedade, Vol. XIII, No. 1: 111–130.
Venturieri, A., A.P.D. Aguiar, A.M.V. Monteiro et al.
(2004). Dinâmica territorial da frente de ocupação
de Sao Félix do Xingu-Iriri: Subsídios para o
desenho de políticas emergenciais de contenção
do desmatamento. Brasília, Secretaria de Políticas
e Programas de Pesquisa e Desenvolvimento:
Rede GEOMA, Ministério da Ciência e Tecnologia.
1. The Amazon Environmental Research
Institute (IPAM), Brazil. The authors are listed in
alphabetic order.
2. Project for Satellite Monitoring of Deforestation
in the Legal Amazon, which has been producing
annual deforestation rates since 1988.
The International Policy Centre for Inclusive Growth
Policy in Focus 35
Marketing of Agroextractive Products:
Problems and Solutions
by Donald Sawyer1
The following analysis and
recommendations are based on the
experience of the Programme of Small
Ecosocial Projects (Programa de Pequenos
Projetos Ecossociais PPP-ECOS) with small
grants to local communities over the past
20 years for projects for the sustainable use
of biodiversity in the Cerrado, Caatinga
and Amazon regions of Brazil.2
The article identies the main problems
and proposes practical solutions, to
support the appropriate design of new
initiatives to promote the marketing of
agroextractive products in frontier regions
in the tropics. Agroextractive activities
include wild harvesting, non-timber forest
products, forest management, agroforestry
systems and small-scale agroindustries.
Under the PPP-ECOS the most successful
experiments so far have been with native
fruits, dried owers, babassu palmnuts,
baru nuts, medicinal herbs, spices,
handicrafts, honey from native and exotic
bees, sh farming and the breeding of
wild animals (ISPN, 2005). Other possible
alternatives that seem promising include
shing, ornamental plants and small
wooden objects, among others.
Producers include small farmers, settlers
in agrarian reform projects, Maroons
(members of former slave communities)
and indigenous peoples.
Using a sustainable livelihoods approach,
the programme seeks to relieve poverty
and improve quality of life through the use
of the environment for social inclusion.
At the same time, alternatives are sought
to the destruction caused by deforestation,
ranching, monoculture, erosion and
pollution. Maintenance of ecosystem
functions related to water, biodiversity and
climate, combined with economic viability
and social justice, requires reaching a scale
far beyond families and communities,
although they are the immediate target.
Marketing of agroextractive products is
generally a seasonal and complementary
activity that is insucient in and of itself
but which makes it possible for people
to stay in the countryside rather than
migrate to cities that opens the way
for cattle ranches and monoculture.
Only peasant macro-landscapes, with
people and production, can produce
development that is economically,
socially and environmentally sustainable
at the system level. To reach that objective,
it is necessary to inuence public
policy, formal and informal educational
systems and the practices of producers,
intermediaries and consumers.
Nearly all the experiences supported
so far continue to generate positive
results. No alternative has been discarded.
On the other hand, none of the
experiences have gained sucient scale to
generate positive impacts on a permanent
basis. A possible exception would be the
production of frozen fruit pulp, although
the existing factories in Minas Gerais,
Maranhão and Mato Grosso still rely on
subsidies which would not be replicable
elsewhere on a larger scale.
Many lessons have been learned from
practical experience. Useful knowledge
is generated not only by researchers but
also by communities themselves,
the support organisations and the
government agencies involved. The main
problems encountered in regards to
practical experience so far relate to:
The lack of scale, quality and regularity:
there are many markets, but not
enough products. While some buyers
want containers delivered every month,
communities can hardly produce
enough for local supermarkets. Due
to natural variations and social
contingencies, it often becomes
dicult to comply with contracts.
Small scale of social and environmental
impacts: domestic production for
self-consumption in local
communities contributes to food
security and nutrition, but the
impacts are very localised.
Amateurism in production and
marketing: processing of products
and interaction with markets beyond
the community level demand
professionalism, not just strong social
or political leadership.
The lack of entrepreneurship: few
smallholders and their leaders have
business acumen. At the same time,
social movements are characterised
by goals and processes that are hardly
compatible with the agile negotiations
required by the market.
Investment requirements: the
formalisation of production and
marketing requires investment in
equipment and qualied personnel,
which also increases the cost of
production. The scale that is possible
with products that are widespread
rather than densely concentrated
rarely justies large investments.
Unfavourable government regulations:
health standards, environmental, tax
and professional regulations favour
conventional industrial and agricultural
production and marketing making
the formalisation of family production
dicult or impossible (Simone, Sawyer
and Almeida, 2011).
Despite good intentions, international
cooperation projects, as well as consultants
with strong social and environmental
commitments, the development of
databases and preparations of business
plans have not solved these problems.
Possible solutions
The practical experience of the PPP-
ECOS over the years suggests that
the problems identied above could
be solved by adopting or adapting the
following approaches:
Support through selling products and
services: the growing shortage of
New initiatives
to promote the
commercialisation of
agroextractive products
should have a strong
focus on the real needs
of peasants and artisans,
ecosystems and
society as a whole.
1. Instituto Sociedade, População e
Natureza (ISPN) and retired Professor,
Center for Sustainable Development,
University of Brasília. The author is grateful
to colleagues for their inputs and comments.
2. This Programme, originally the Brazilian
version of the Small Grants Programme of
the Global Environment Facility and UNDP,
also receives or received support from the
European Commission and the Brazilian Banco
Nacional de Desenvolvimento Econômico
e Social for the arc of deforestation in the
Amazon. None of the donors are responsible
for the views expressed herein.
donor resources and the diculty
of spending and accounting for
government funds according to
existing rules and regulations mean
that community organisations will
have to sustain themselves, at least
partially, by selling products and
services and participating in markets,
even when they are non-prot.
Need for subsidies and incentives: some
subsidies and incentives, be they
direct or indirect, explicit or not, from
government or donors, are required for
agroextractive enterprises, especially
in the investment phase, and are
justied because of environmental
multifunctionality that generates
benets at a macro scale.
Appropriate level of processing: since
nal processing for consumers
in accordance with government
regulations often requires operations
at a large scale, local initiatives
should seek an intermediate level of
processing—one that is feasible given
the availability of infrastructure as well
as human and nancial resources.
Prioritising nearby markets: the markets
to be targeted should start at the local
level and proceed to regional, national
and international levels, in that order.
Entrance to distant markets, when
feasible, requires improvement in
quality and provides greater visibility.
Search for opportunities in normal trade:
apart from fair or solidary trade, which
can occupy certain signicant niches
for small producers, local communities
should seek inclusion in increasingly
larger segments of conventional
markets, so as to achieve sucient scale.
Training and guidance: buyers of
agroextractive products should
provide feedback for producers
regarding appropriate procedures,
consumers’ perceptions and
regulators’ requirements.
Diversification and specialisation: each
community should seek to combine
a set of three to ve complementary
yet diverse products, without
excessive variety, but also without
accumulating risks.
Producer organisation: the social
organisation of producers is essential
for greater bargaining power in
negotiating with buyers and social
control to avoid abuses by uninformed
or irresponsible individual producers.
However, perfect unity should not be
expected, since communities always
have internal divisions. Associations
should not be imposed.
Collective bargaining with external
support: local social organisations
should seek external support from
regional or national organisations
that have greater bargaining power
with buyers.
Recognition of mutual benets: buyers
should recognise the competitive
advantages of oering agroextractive
products to consumers, without giving
the impression that they are engaging
in philanthropy.
Organisation of buyers: the
organisation of buyers of
agroextractive products, adhering to
ethical principles and sustainability,
is important to ensure social control
within the business sector.
Exemptions for handmade products:
original peasant and artisan products
on a small scale, should be exempt from
regulations that apply to production
and marketing on a large scale.
New initiatives to promote the
commercialisation of agroextractive
products should have a strong focus on
the real needs of peasants and artisans,
ecosystems and society as a whole. With
appropriate policies, agroextractive
production can meet those needs at low
costs. The local communities require
cooperation and support, without
unrealistic impositions that could defeat
their purpose.
ISPN (2011). Long live the Cerrado! Sustainable
products and livelihoods supported by the GEF
Small Grants Program in Brazil. Brasília, Instituto
Sociedade, População e Natureza.
Simone, J., D. Sawyer and F.V.R. Almeida
(2012). Entraves regulatórios na produção
agroextrativista. Brasília, Instituto Sociedade,
População e Natureza, <
(accessed 25 March 2014).
The International Policy Centre for Inclusive Growth
Policy in Focus 37
Modelling of Deforestation Scenarios
for the Northwest of Mato Grosso
by Britaldo Soares-Filho1 and Raoni Rajão2
Source: Authors’ elaboration.
The northwestern region of Mato
Grosso, located 740 km from Cuiabá,
takes up 112,000 km2 and encompasses
the municipalities of Aripuanã, Colniza,
Cotriguaçú, Juruena, Castanheira,
Rondolândia and parts of Brasnorte and
Juará. The city of Juína is the region’s main
economic hub. This region has attracted
attention for two contrasting aspects.
First, it is the nal border of the Amazon
rainforest in the state of Mato Grosso, with
rainforest covering 87 per cent of its area.
Of this total, 93,000 km2 is covered by
dense and open rainforests (MME, 1973),
making the region particularly relevant
both for conservation purposes and for
the economic use of its signicant forest
resources. To this end, the region has been
the target of eorts to create conservation
units and consolidate its vast network of
indigenous reserves, occupied by diverse
peoples and cultures and comprising 34
per cent of the region.
Second, despite its scarce population of
about 100,000 inhabitants, the region
is notable as one of the main areas of
deforestation and associated conicts,
such as illegal logging and rural murders,
which confer on Colniza the sad title of the
Brazilian city with most murders per capita.
The municipality was also negatively
highlighted recently as one of the main
areas of deforestation in the Amazon
(INPE, 2008). This fact is well illustrated
by the analysis of the spatial dynamics of
deforestation in Mato Grosso, which shows
the shifting of the deforestation frontier
towards this region in recent years.
This article presents the main results of
a study conducted in the context of a
United Nations Development Programme
(UNDP) and Global Environmental Fund
(GEF) programme for conservation and
sustainable use of biodiversity in forests in
the northwest of Mato Grosso.
In particular, this article presents an analysis
of the recent dynamics of deforestation in
the northwest, and calculates modelled
deforestation trajectories in the region until
2035 under trend, business-as-usual and
governance scenarios.
Deforestation scenarios
A rst step in the development of
deforestation trajectory modelling is
an analysis of its historical trend. To this
end, data from the PRODES3 time series
of satellite monitoring (INPE, 2008) were
crossed-referenced with data from the
Secretary of the Environment of the State
of Mato Grosso (SEMA).
Accumulated deforestation up to 2005
represents a total of 13 per cent of the
northwest region’s 112,000 km2—or
21 per cent if the protected areas that
comprise 40 per cent of the region are
subtracted from this total. Both series
highlight an acceleration of deforestation
at the beginning of this decade, reaching
a maximum of 300,000 hectares per year
in 2004/2005, according to PRODES, or
180,000 hectares per year according
to SEMA (INPE, 2008). However, in
2005/2006, there was a sharp drop in the
deforestation rate in northwestern Mato
Grosso. The causes of this decline are
still under discussion but are generally
recognised as a combination
of two factors:
a drop in soybean production due
to the appreciation of the Brazilian
Real and a reduction in the
international market price for soybeans,
which had an impact on agricultural
investments, including not only
agriculture but also livestock. This
resulted in a decrease in the opening
up of new areas. However, this trend
was reversed after 2007; and
EGO computing platform (Rodrigues et al.,
2007; Soares-Filho et al. 2013).
Based on the results of the model, very
dierent trajectories for these three
scenarios can be observed. In the BAU
scenario there would be an expansion of
agricultural areas that presently occupy
about 30,000 hectares (300 km2) in the
region to more than one million hectares
(10,000 km2), with an average annual
growth of 14.4 per cent, essentially leading
to the occupation of the entire area t for
mechanised agriculture in the region.
On the other hand, in the GOV scenario the
cattle herd would expand by 2.1 per cent
per year, resulting in 3.1 million cattle by
2035. The agricultural areas would reach
85,000 hectares (850 km2) at an expansion
rate of 4.4 per cent per year. Thus, in the
TREND scenario there is a slight decrease
in annual gross deforestation due to the
xed annual net rate. In the GOV scenario,
deforestation drops to almost zero
at the end of 2013. In contrast, in the
BAU scenario deforestation increases
and may even exceed 200,000 hectares
(2,000 km2) per year.
This gure was surpassed in the recent
past, when it reached 300,000 hectares
(3,000 km2) annually between 2004
and 2005, according to data from
PRODES (INPE, 2008). While the GOV
scenario forecasts a further increase of
around 500,000 hectares (5,000 km2)
of deforestation, the increase will be four
million hectares (40,000 km2) in the trend
scenario and 5,800,000 hectares (58,000
km2) in the BAU scenario. The two latter
scenarios involve, respectively, 33 per
cent and 52 per cent reductions from
the current 9,300,00 hectares
(93,000 km2) of the region’s forests.
The prevalence of either the trend scenario
or the BAU scenario for the region will have
devastating consequences, with almost
complete extermination of forests outside
protected areas (Map 1). Yet the trend
scenario was considered conservative,
since it uses the 2005 deforestation rate,
which is much lower than that of more
recent years. If the BAU scenario prevails—
which has in fact been happening in
the recent past—protected areas will
substantially lose their forest cover,
especially the Escondido, Jaruíra, Arara
do Rio Branco, Serra Morena and Sete de
the increased ght against
deforestation and illegal logging by
the State, with the launch of several
eld campaigns in conict areas and
police operations against corruption
networks (e.g. Operation Curupira).
In terms of possible future deforestation
trends, three scenarios were modeled,
namely: a) trend (TREND); b) business-as-
usual (BAU); and c) governance (GOV).
The TREND scenario is known as such
because it employs the historical rate,
estimated by SEMA for the year 2005/2006,
to project future deforestation in the next
30 years—i.e. around 125,000 hectares
per year or equivalent to the net rate of
1.33 per cent per year.
This scenario can be regarded
as conservative, given the more recent
trend of accelerating deforestation.
The TREND scenario was calibrated for the
northwest region using a series of SEMA
deforestation maps with a set of variables
used to represent the spatial determinants
of deforestation—for example: altitude,
distance to roads, rivers, urban patches
and previous deforestation, the historical
eect of protected areas and classes of
licenses, slopes and soil conditions.
The methodology used is described
in Soares-Filho et al. (2006). In short, it
consists, as a rst step, of calibrating
the eect of these determinants on
the spatial allocation of deforestation,
using the weight of evidence method.
Then, as a second step, the architecture
of the simulation model was adjusted
to the resolution of one hectare and
The region
[Mato Grosso] has
been the target
of efforts to create
conservation units
and consolidate
its vast network
of indigenous
reserves, occupied
by diverse peoples
and cultures
and comprising
34 per cent
of the region.
set up to reproduce the spatial pattern
resulting from the arrangement of the
dierent actors—large farmers and
small settlers—and distributing the
historical deforestation rate according
to their occupation of the landscape and
adjusting the transition functions to form
patches with sizes ranging, on average,
from 10 hectares to over 100 hectares.
One can thus say that deforestation
in the region is a combined process of
small and large farmers, with the former
occupying 55 per cent and the latter
45 per cent of the area already cleared
outside the protected areas. As a result,
while small settlers deforest areas smaller
than 10 hectares per year, large farmers
can deforest areas of more than 100
hectares in a single year.
Deforestation trajectories under the BAU
and GOV scenarios were dened based on
the modelling architecture described in
Soares-Filho et al. (2008), responding to the