Previous PageTable Of ContentsNext Page


3.1 Legal Amazon – definition and forest cover

Located in South America, Brazil has a territorial surface of 851 million ha, of which approximately 560 million ha are covered by native forests and other wooded vegetation. Of this total, more than two-thirds are formed by the Amazonian tropical rain forest and the rest by transition forests, cerrados, caatinga, Atlantic coast forest and its associated ecosystems (see Box 1, Brazilian and Amazonian Ecosystems)

From a political-administrative point of view, the country is divided into five big geographical regions (Table 3.1). The northern region, which has the largest surface, occupies an area of more than 358 million ha. Together with the centre-west region, both occupy 64 per cent of Brazilian territory. Notwithstanding this, the population of this enormous territorial space corresponds to only 15 per cent of the total population of the country. The high population concentration in Brazil occurs in the northeast and southeast regions, that together make up 29 per cent of the national territory, and where nowadays 71 per cent of Brazilians live.

Table 3.1: Brazil – Area and population per region – 1996 and 2000


Area (ha)

% of Brazil

Population 1996 (inhabitants)

Population-2000 (inhabitants)

% of Brazil





































Source: IBGE, 1996; IBGE – Censo de População 2000 in STCP

Two geographical concepts of Amazonia are used in Brazil - Classical Amazonia and Legal Amazonia (see Map 3.1). Classical Amazonia corresponds to the northern region of Brazil and includes seven states (Acré, Amapá, Amazonas, Pará, Rondônia, Roraima and Tocantins), the total area of which is 120 million ha smaller than Legal Amazonia.

The Brazilian Legal Amazon corresponds to a region legally defined for purposes of regional planning and public policy with a total area of 510 million ha (an area that is sufficiently large to accommodate all of western Europe), encompassing all of the northern region and parts of the centre-west and northeastern regions. Legal Amazonia includes a total of seven states (Acré, Amapá, Amazonas, Pará, Rondõnia, Roraima and Tocantins) and parts of two other states (Maranhão and a very small part of Goiás State, which is not included in Table 3.2).

The total surface of forests and other forested-land of Brazil is of the order of 560 million ha, and is equivalent to 14.5 per cent of the world forest cover. While the world average per person of forest surface is 0.6 ha, with great variations per country, the Brazilian average reaches 3.3 ha per person (2000) but also shows strong variations according to the region of the country. The crossing of the data on total forest area of the states of the northern region (Classical Amazonia), cf. Table 3.2, with the data on regional population (Table 3.1, for 1996) shows that the north has a per capita rate of forest surface of 30.3 ha, which is almost tenfold the national average and 50 times bigger than the world average.

Table 3.2: Brazilian Legal Amazonia – surface (ha) and forest cover






Original area of forests





Original area of cerrados




Other biomas








- 0 -










- 0 -


















Mato Grosso






































- 0 -










Sources: (1) IBGE, 1996 ; (2) INPE, Desflorestamento 1995-97: it refers to forest physiognomies - dense and open rain forests, transition forests, cerrados, areas of contact, seasonal forests and wooded savannas – data adapted by the author from Nepstad, D. et al., (3) data adapted by the author from Andrade, E. B. and from Menezes, M.A.; (4) obtained from residue.

According to Funatura1 and Schubart2, the forest cover types of the almost 398 million ha of the Amazonian tropical forest are divided into 79 per cent of dense non-flooded tropical rain forests (terra firme forests), 10 per cent of open non-flooded tropical rain forests (terra firme forests), 2 per cent of flooded tropical rain forests (varzea forests, periodically flooded, and igapó forests, permanently flooded), 6 per cent of open vegetation cover types (such as the savannas, varzea fields and campinaranas) and the rest (almost 3 per cent) constituted by aquatic vegetation, rivers and lakes.

Therefore, the dense and open tropical rain forests, flooded and non-flooded, constitute the most representative forest cover of Legal Amazonia and are equivalent in area to 30 per cent of the tropical rain forests of the world, besides being considered as the richest ecosystem of the earth in terms of bio-diversity.

3.2 Population dynamics

Based on Sawyer, D.4, and on the wide-ranging analytical work carried out by the Society, Population and Nature Institute (ISPN) about demography and quality of life in Amazonia, it is possible to summarize the main characteristics and trends of the population dynamics of the region that have major relevance for the understanding of the association of such dynamics with the process of deforestation in the region, in the last three decades (1970-1996).

Composing the framework of the spacial distribution of population in the region, which may be seen in maps 3.2 and 3.3, five main trends, as put forward by the cited author, describe the dynamics of the population in Amazonia in the last 30 years, and must influence future outcomings of the complex relationship between population and deforestation in the region:

Map 3.1: Brazilian Amazonia, classical Amazonia (north region) and Legal Amazonia

Box 1

Brazilian and Amazonian ecosystems

As per the study carried out by the Ministry of the Environment, “Brazilian Ecosystems and the Main Macro-Vectors of Development – Subsidies to the Planning of Environmental Management”, in 19953, the country may be divided into nine major sets of ecosystems, understood as territorial macro-units for analysis and planning, and which aggregate certain basic characteristics that constitute a background for environmental management.

Given the Brazilian environmental complexity, the nine sub-spaces were defined, basically, according to ecological conditions reflected by the predominant original vegetation and by the geographical position that they occupy, except the Pantanal, a region that has its own geo-morphological characteristics. The nine ecosystems were defined independently of the political-administrative division and according to regionalization criteria already employed. The nine units were denominated relative to the geographical and/or phyto-ecological region, always preceded by the word “ecosystem”, considering ecosystem as a generic term with respect to the diverse levels of organization of the biosphere.

The nine main Brazilian ecosystems are the following: 1. The Amazonian ecosystems; 2. The ecosystems of the cerrados region; 3. The ecosystems of the Pantanal; 4. The ecosystems of the caatingas region and of decidual forests of the northeast; 5. The ecosystems of the middle-north; 6. The ecosystems of the region of semi-decidual seasonal forests; 7. The ecosystems of the Brazilian pine region; 8. The ecosystems of the extreme south; 9. The ecosystems of the coastal regions and of the Atlantic forest.

We present below only a brief description of the Amazonian ecosystems and of the ecosystems of the Cerrados region, since they are the main ones encompassed by the region defined as “Legal Amazon”, the main object of this work.

1. Amazonian ecosystems: the Amazonian ecosystems are dominated by the tropical Perenifolia or rain forest, which represents the Brazilian portion of the “Hylea” of Humboldt and Bompland. They still constitute the largest body of tropical forest of the planet, conditioned by a hot equatorial humid and super-humid climate, where rain precipitation is far superior to potential evapo-transpiration.

Three big morphological sets, related to regional geology, compose the relief of Amazonian ecosystems: in the centre, the interior planes and the central Amazon depression; to the south, the meridional Amazon, the Araguaia-Tocantins river depressions, the plateaus of the Tapajós-Xingu rivers and of the Parecis, the residual plateaus of oriental Amazon and meridional Amazon; to the north, the northern Amazon depression and the plateaus of Amazonas-Orinoco and Negro-Jari rivers, as well as the residual plateaus of northern Amazon.

The great geological diversity of the Amazonian ecosystems, linked to the differentiated relief, resulted in the formation of various classes of soils, with emphasis on alluvial and hydro-morphic soils, yellow-latossol and red-yellow latossol, and podzolic soils.

The dense rain forests, represented/characterized by terra firme forests, várzea forests (periodically flooded) and Igapó forests (permanently flooded) represent the core area of the “Hylea”, ranging from the northern Amazonian depression, a major part of the Amazonas-Orinoco plateau, and for almost all of central-Amazonian depression to east of Meridian 70.

The open rain forests characterize an area of climatic transition, less humid, and are also known as terra firme forests, várzea forests and Igapó forests, according to the topographic position that they occupy and the river geo-morphology aspects. They occupy a major part of the southern Amazonian and the Araguaia-Tocantins depressions, as well as the drained areas by the courses of the Javari, Juruá, Purus and Madeira rivers.

The várzea fields of the Amazon river, from the merging of the Solimões and Purus, to the islands of its delta, have special importance. The natural fields of Roraima (known as Esteppe-Savannas) occur in the southern extreme of the Branco and Tacutu river basins. The campinarana, the vegetation of which defines a specific phyto-ecological region, spreads in spots along the basin of the high-medium Negro river. The transition area mainly characterizes the strip near pre-Amazonia. The Amazonian ecosystems occupy a surface area of 400,508,200 ha. It is estimated that the waters and native vegetation still cover 90 per cent of the whole area of these ecosystems.

2. Ecosystems of the cerrados region: the cerrado is the Brazilian variant of the class of vegetation formation originally called savanna by Olviedo, when he characterized the Venezuelan “llanos”. Ecologically speaking, the Brazilian cerrados characterize a set of environmental factors that confer individuality to central Brazil.

The interaction of morpho-climatic, structural and lythological factors results in the compartmentalization of this area into big geo-morphological units, distributed in differentiated altitude levels: chapadas, plateaus and layers elaborated in diverse erosive phases can be found separate from the high courses of the rivers that form the basins of the Amazon, of the Sâo-Francisco and of the Prata, originating in the Araguaia-Tocantins, the high and medium São-Francisco and the high Paraguay depressions.

The dominant soils of the plane and smooth-ondulated areas of the chapadas and plateaus are the dark-red latossols, the concretionary soils and quartz sands. In the mountainous and/or desicated areas, there are the podzolic, cambissols and litolic soils. The plyntossols and gleys are found in the depressed areas subject to periodical flooding.

The vegetation of the cerrados – latu sensu – presents a varied physiognomy. The Cerradão is a dense tree-formation. The cerrado – strictu sensu – is constituted by tortuous trees, relatively short, intertwined by bushes. In the cerrado fields, the trees and bushes are spaced. The clean fields, or cerrado fields, occur in depressed, badly drained areas, and are poorly endowed with woody vegetation.

The ecosystems of the cerrados region occupy a surface of 189,027,800 ha. It is estimated that the natural vegetation of cerrados, fields and forests represents 76 per cent of the total area of these ecosystems, although a major part of the remainder are already altered. The table below summarizes the main vegetation types of the nine groups of Brazilian ecosystems mentioned here.

Main Brazilian ecosystem and vegetation types


Vegetation types

Original area (ha)

% of total country area


Dense and open rain forests, várzea fields, esteppe savanna (Roraima fields), campinaranas, palm-tree forests, bamboo forests, transition areas and others.




Cerradão, cerrado, cerrado-fields, clean fields, river-margin woods and veredas




Cerrados, cerrado-fields, clean fields, chacos




Caatinga arboreal, caatinga bush-arboreal, bush caatinga, decidual forests




Dense rain forest, seasonal forests, contacts with cerrado and seasonal forests, contacts with cerrado and caatinga



Seasonal semi-decidual forests

Seasonal semi-decidual forests, transition areas with decidual forests, Atlantic forest and cerrado.



Brazilian pine forests

Mixed rain forest or Araucária forests, grass fields




Fields, savannas, seasonal forests, esteppes and esteppe-savannas



Coastal areas and Atlantic forest

Mangroves, coastal fields, dense and open rain forests, altitude fields and others.







Incomplete demographic transition: in spite of having fallen (from 7.9 children per woman in 1970, to 4.2 in 1991), the total fecundity rate of the Amazon continues to be the highest in Brazil (the average of the country was 2.9 in 1991). On the other hand, regional life expectancy increased from 53.8 years in 1970 to 68.4 years in 1991; a bigger increase than the country’s life expectancy as a whole, which grew from 53.7 years in 1970 to 66.3 in 1991. The lower mortality rate in the Amazon is due to such factors as reduced demographic density, direct access to abundant natural resources as a means of survival, abundance of water, etc. The combination of high fecundity and lower mortality than the national averages means that the vegetative growth rate of the Amazonian population continues to be relatively high (2 per cent per annum, compared to 1.5 per cent for Brazil as a whole).

Decrease of inter-regional migration: until the mid-1980s, migrations from other regions to the Amazon were relatively intense, but even so the net migration surpluses to the region contributed less than the vegetative growth of the population, (migration flows in the 1970s and 1980s represented 35 per cent and 41 per cent, respectively, of total regional population growth in these periods). The migration surpluses decreased significantly in the second half of the 1980s, and the beginning of the 1990s. The reasons for the fall in inter-regional migration to the Amazon are linked to the suspension of big colonization projects and public construction works, to the tightening of land conflicts, disillusion with access to lands, malaria epidemics, scarcity of capital to invest in migration, etc.

Increase of intra-regional mobility: nowadays, instead of migration from other regions to Amazonia, there is great population mobility inside the region.

This intra-regional mobility is destined for urban areas of all sizes, placer-minings, new settlements (mainly in Roraima, in the north of Mato Grosso and south of Amazonas) and is due, in part, to the instability of existing settlements and to the search for survival opportunities, as well as being connected to high demographic growth in the areas of origin. Associated with this intra-regional migration there is the temporary mobility of labour (mainly male) due to the seasonality of agriculture production, cattle-ranching and extractivism, to the regional calendar of deforestation tasks, to civil construction works and other activities. The temporary mobility has been facilitated by improvements in the systems of infrastructure and transportation.

Accelerated and generalized urbanization: one of the most outstanding characteristics of Amazonia, nowadays, is the occurrence of an accelerated and generalized urbanization of the agriculture frontier, the population of which is around 60 per cent urban and continues to increase. The urban population of Legal Amazonia is today 12 million people – which means a great consumer market that demands agricultural products, industrialized products and urban services – and which is concentrated in a few big cities (above 500 thousand inhabitants – Belém, Manaus, Cuiabá, São Luís), and in some tens of small and medium cities (from 20 to 250 thousand inhabitants), mainly in the region of the deforestation arch, in the northeast of Pará, in the centre of Maranhão, along the Amazon river and in the region of Manaus.

Map 3.2: Total demographic density – 1996

Map 3.3: Total population – 1996

Sources: MMA, “Causas e dinâmica do desmatamento na Amazonia”

The fixation of rural population: notwithstanding the intense urbanization of Legal Amazonia, around 8 million inhabitants still remain in the rural areas of the region, a considerable population that keeps on growing

Maps 3.4 and 3.5 show that rural population is more concentrated where urban population is concentrated. It should be noted that there still is a reasonable diversity in the composition of this rural population in its distribution in the region: family producers (peasantry, shifting cultivators, based on household labour), farmers and rural producers (forms of production that rely on the hiring of waged labour) are concentrated in Maranhão and in all of the deforestation arch region, while the so-called traditional/extractivist populations, like the rubber-tappers, Brazil-nut gatherers, fishermen, riverains and others are found in more remote areas. There still are around 300,000 indigenous people (a population that is growing in the Amazon) that occupy a total area of more than 100 million ha (20 per cent of Legal Amazonia) in reserves distributed all over the region.

The relationships between population and deforestation are complex. Examples of this complexity in the Amazon are: (i) in Amazonia, for example, demographic density does not correspond to anthropogenic pressure in a linear form: the rural exodus of low-environmental impact extractivist or agricultural populations may point to risks of introducing predatory agricultural or timber exploitation activities; the fixation of rural population means that, contrary to the past, many rural producers are interested in the sustainability of their lands, and so it is less likely that they move on to new frontiers opened by timbermen if they are displaced by the expansion of cattle-ranching or soya bean crops; (ii) the urban population also interacts in a complex form with deforestation: a) on the one hand, migration to cities diminishes the direct pressure on the forest (in Amazonas, the concentration of population in Manaus – because of the free-trade zone – contributes to the low level of deforestation in the state; b) on the other, the growth of cities creates a demand for food, raw material, energy and transport, that stimulate deforestation.

Map 3.4: Urban population – 1996

Map 3.5: Rural population – 1996

Source: MMA, “Causas e dinâmica do desmatamento na Amazônia”

Finally, although population growth exerts a certain pressure on deforestation, the effects are not always direct and linear. The effects of regional population increase on deforestation cannot be seen as Malthusian (cf. Margulis and Reis5).

3.3 Soils, climate and present use of land in Legal Amazonia

3.3.1 Soils

A good part of the preoccupation, at national and world levels, about the deforestation process under way in the Amazon is due to ecological and environmental consequences, present and potential, that result from the use of the region’s natural resources. In this context, for the last three or four decades, the development of agriculture and cattle-raising, as practised in Amazonia, has been one of the most important factors that determine environmental disturbances and impacts in the region, to a great extent because the expansion of these activities, carried out through clear-cutting and conversion of forests for the establishment of crops and pastures, has been an important, if not the major, determining factor of deforestation in the region.

Amid this process, the question of whether the agricultural development of the Amazon can be sustainable or not has been widely debated, and is still a controversial issue, particularly in respect of the aptitude or agricultural potential of the region’s soils.

Although part of the controversy is due to the scarcity of more detailed knowledge on the types of soils, and their real extensions in the immense region, and without the pretension to exhaust the theme, one may say that only one statement seems to have unanimity among the specialists on tropical soils: that only 8 to 10 per cent of the Amazon’s soils have high natural fertility (around 40 million ha, constituted of eutrophic soils, that do not show any limitation in the physical or chemical structures – see map 3.6).

For the rest, the opinions are contradictory or polarized. For example, in a recent study sponsored by the World Bank6, the authors state: “there is ample literature highlighting the low agricultural potential of the most part of the Amazon (Goodland and Irwin, 1975; Moran, 1981; Smith, 1981; Cochrane and Sanchez, 1982; Hecht et al., 1988; Mattos and Uhl, 1994 ). Schubart (1999), for example, concludes that approximately 90 per cent of Amazonian soils are acid, chemically poor and have excessive humidity, which favours in a significant way the development of disease and plague...” (our underscoring).

On the other hand, authors like Andrade, E.7 state that: “from the edaphic point of view, the Amazonian soils in their great majority (56 per cent) have high agriculture aptitude, and, maintaining the adequate management indexes, they can show very good productivity for crops and cattle-raising. Approximately 36 per cent of some classes of soils that show an unbalanced physical structure and low rates of base-saturation are inadequate for agricultural practice. Around 8 per cent of the soils (40 million ha) are made of eutrophic soils, that do not show any limitation in the physical or chemical structure, but generally are located in areas with easy relief, which makes mechanization difficult, and part of them are located in zones where transportation means are precarious (Silva, 1981 and Alvin, 1994)” (our underscoring).

Map 3.6: Distribution of the main classes of soil in the Amazon

Yet, according to this author, “in all states of the region soils with good physical structure predominate, and although they may have low natural fertility, if deficiencies are corrected with the specific fertilizers, the productivity of the crops has been showing satisfactory economic results. Independently of the use to which the soils of the Amazon are destined, the sustainability of the production systems will be conditioned according to the employment of efficient conservation practices against erosion, mainly in the more mountainous areas and to adequate processes of management, where the use of coverage plants is important for the success of the agricultural projects” (our underscoring).

3.3.2 Climate

More recent scientific discussions and evidence have emphasized the role of climate and its influences in the determination of agricultural production in the Amazon. The quoted study of the World Bank6 mentions the recent work of Chomitz and Thomas8, in which the authors verified that pluviosity has a negative effect on agricultural productivity in Amazonia. The two authors attempt to show that the influence of infrastructure (road network, markets, communications) on the increase in agriculture productivity in Amazonia is relative: as per their analyses, depending on the activity being located in a high rainfall zone, the productivity and permanence of crops may not respond to the existence of infrastructure, i.e., infrastructure will not guarantee the economic feasibility of the crops. Chomitz and Thomas identified three major rainfall zones with significant influences on the performance of agriculture in the Amazon (see map 3.7).

The pattern of humidity and water availability in Legal Amazonia tends to a gradient in which rainfall, from 1,000 mm in the south, grows in a northwest and northeast direction, reaching annual averages up to 3,000 mm. In a general way, the Amazon region is characterized by the following basic climatic components: high precipitation that features a rainy season, with an average of 2,000 mm/year, that is extended for 180 days; high relative humidity of the air that marks this rainy period; and monthly average temperature, around 26°C, without variation higher than three degrees among the different months of the year.

Map 3.7: Main rainfall zones of Amazonia

According to Andrade, E. (op.cit), in the more humid areas of Amazonia, the development of a more technified agriculture, and on an entrepreneurial scale, will only be profitable if infrastructural and logistical support conditions, capable of offering comparative and competitive advantages, are available in order to raise investors’ interest. Besides these structural factors, the high humidity and uniform heat during the whole year propitiate the dissemination of pathogenic biotic agents that may create serious obstacles to economic sustainability of the production systems, due to additional costs of the intense and necessary utilization of pesticides and defensive products.

In this sense, the natural vocation of these more humid areas seems to be the exploitation of native forests through sustainable systems and the use of bio-diversity or perennial crops with a high degree of technological adaptation (as is the case of palm-oil, açaí or cocoa crops, among others).

In the strips at the south of the Amazon basin, more in contact with the central plateau, due to the influence of cerrados (a region that corresponds to the deforestation arch and to the penetration of the agricultural frontier) and the drier climatic conditions, the crops seem to present better performance in the face of predator biotic agent actions. In general, there are no grave climatic limitations related to availability of light, water and thermal amplitude capable of rendering infeasible systems of production for those products that already have technological innovations in their production processes.

Box 2

Rainfall zones of Legal Amazonia

Dry Amazonia – rainfall below 1,800 mm/year: this area corresponds approximately to 17 per cent of the territory of the region. It is concentrated at the south of the Amazon Basin (deforestation arch) and in isolated areas of natural fields mainly in the north of Roraima. In this region climatic conditions are relatively favourable to agriculture. The soils are generally well drained, and relief is favourable to mechanized agriculture. Vegetation is mainly cerrado, with some areas of open and semi-deciduous forests, which have low volumes of commercial wood species.

Transition Amazonia – rainfall between 1,800 mm and 2,000 mm/year: this area represents approximately 38 per cent of Amazonia and is located between the central region (humid zone) and the deforestation arch in the south of Amazonia (dry zone). This region is generally covered by dense forests with zones of open forests in Mato Grosso and south of Pará. The soils are reasonably well drained. The relief is mostly ondulated with some significant elevations in Roraima and north of Pará. The excess of rain and short dry period create agricultural and economic difficulties for the production of grains. Perennial crops have had better success. In the case of cattle-raising there is some evidence of relative success for more intensive technology of production in eastern Pará (Paragominas).

Humid Amazon – rainfall higher than 2,200mm/year: there are some areas in this zone that register rainfall up to 4,000 mm/year. In many areas soils have insufficient drainage. This zone, embracing 45 percent of the Amazon, is located mainly in the central region, occupying most parts of the states of Amazonas and Amapá, the northwest of Rondônia, the southwest and northwest of Pará, besides the Marajó island and Bragantina region (northeast of Pará). Dense forests cover the biggest part of this zone. The excess of rain and insufficient drainage tend to make agriculture (especially the cultivation of grains) economically non-competitive. The profitable activities occur only in areas with better infrastructure and market. This big area, around 84 percent, has average or high potential for timber exploitation, according to data from IMAZON9.

3.3.3 Present use of land The growth of agricultural activity in the Amazon

The economic growth of the northern region of Brazil, in the last four decades, has been quite intense. Throughout this period, the Amazon has shown a higher rate of economic expansion, compared to the other Brazilian regions, whether measured by total GDP or by GDP per capita. From 1960 to 1996, while the country grew at an average annual (a.a.) rate of 5.2 per cent, the north presented a more expressive performance, of 8.6 per cent a.a. (Table 3.3) with the post-1970 period showing a more accelerated growth. This caused Amazon GDP to become 7.0 per cent of the country’s GDP in 1996, as against only 3.5 per cent in 197010.

Legal Amazonia GDP per inhabitant, in spite of the significant growth in regional population in the 1970-1996 period, expanded from US$ 1,100 (1970) to US$ 2,300 (1980) and reached US$ 2,850 in 1996. This economic expansion placed the region’s GDP per capita at 58 per cent of Brazil’s per capita GDP in 1996, when it was only 41 per cent in 1970.

In terms of sectoral composition of the regional GDP (see figure 3.1), it can be verified that in the Amazon, as in the rest of the country, the mark of economic growth has also been urban--industrial, that is, the services (including government) and industry (including mining) sectors have shown greater expansion of their relative positions in GDP. The urban-based services sector dominates the Amazon economy.

Table 3.3: Legal Amazonia – GDP, per capita GDP and share in Brazilian GDP





Amazonian GDP (in US$ 1,000 of 1998)




Amazonian GDP per capita (in US$ 1.00 of 1998)




% of Amazonian GDP in Brazil’s GDP




Annual GDP rate of growth – Amazonia (1960/1996) : 8.6%

Annual GDP rate of growth – Brazil (1960/1996) : 5.2%

Source: Monteiro Neto, A.

In the 1970s and 1980s the mining industry has shown the highest growth in the Amazon region. Mining puts pressure on the forests either from its impact on local demand for agricultural goods, or because of its direct environmental effects. The mineral wealth of Brazilian Amazonia has been estimated to be at least US$ 3 trillion, with deposits of gold, bauxite, iron, copper, uranium, potassium, niobium, manganese, diamonds and other precious stones. New mineral deposits are discovered every year. Some mineral reserves are yet to be exploited, such as the gigantic copper reserve of Salobo (in the Carajás area, in Pará), with an estimated stock of 1.2 billion tonnes of copper.

The bigger expression of industry and services in the region’s GDP, in the 1970-1996 period, indicated a reduction in the relative participation of the agricultural sector. Nevertheless, while the relative participation of agriculture and cattle-raising in GDP is reduced, its growth is stimulated in absolute terms, since wider consumer markets for agricultural products are created in the local urban areas with higher income levels. As commonly pointed out, agriculture in Amazonia seems to depend a lot on the existence and availability of local markets, given the long distances and high transportation costs for agricultural products to the other regions of the country.

As some authors single out (Schneider, R.6, for example), the rapid growth of urban markets in Amazonia has been crucial to the feasibility of agriculture in the region. Rapid urbanization in Legal Amazonia points in the direction of a clear modification in the terms of trade between the rural and urban sectors, in favour of the agricultural producer, thus contributing to an increasing profitability of agriculture in the region.

Figure 3.1: Legal Amazonia – sectoral composition of regional product – 1970-1996

Source: Monteiro Neto, A.

So, notwithstanding the fall in the relative participation of agricultural activities in the regional total, from 29.8 per cent in 1970 to 17.1 per cent in 1996, the absolute value of agriculture GDP grew more than threefold, from US$ 2.5 billion (at 1998 prices) in 1970 to US$ 9.1 billion (at 1998 prices) in 1996. Legal Amazonia – altered areas by the expansion of agricultural activities

An idea of what represents the expansion of agriculture and cattle-raising in Legal Amazonia, in terms of a global alteration of the natural landscape of the region, may be obtained from the data given in Table 3.4. The INPE data on gross deforestation occurring in the Amazon reveal that, up to August 1996, more than 51 million ha of forests had been converted, around 13 per cent of the region’s forest cover. The four states that mostly lost forest areas, in absolute terms of area extension, have been Pará, Mato Grosso, Maranhão and Rondônia.

In relative terms of their own surfaces of original forest area, Tocantins was the one that mostly converted forests to alternative land uses (84 per cent), followed by Maranhão (68 per cent), Mato Grosso (28 per cent) and Rondônia (23 per cent).

An approximate estimate of the extension of the modification of vegetation cover of cerrados and other biomas, provoked by anthropogenic action in Legal Amazonia, may be obtained if we compare the data on total altered area in the region, as provided by the 1996 agricultural census of IBGE, with gross deforestation of INPE (the definitions of IBGE on altered areas may be found in notes to Table 3.4).

Compared to the INPE statistics, and given the adopted definitions and different methodologies of getting the information, the data on altered area of IBGE census are consistently higher than those of INPE (except in the case of Acré, where the area of cerrados and other biomass, relatively insignificant, would normally not have alterations usually captured by IBGE) revealing that anthropogenic action and the use of land in Amazonia also encroach, and quite intensely, into areas previously occupied by the vegetation of cerrados and other biomass.

In the Amazon as a whole, the estimates of altered areas of cerrados and other biomas, as presented in Table 3.4, show that until the end of 1996 around 26 million ha of these types of vegetation would have been converted to agricultural use. Although this represents only half of what has been deforested, in absolute terms in forest areas, in relative terms (proportion of the area of each bioma), however, the cerrados are the ecosystems mostly affected: 23 per cent of the area already converted, against 13 per cent in the case of forests, in all of Legal Amazonia. As the cerrados of Legal Amazonia are more representative in the so-called deforestation arch (see below), the expansion of the agriculture frontier to the south of the Amazon basin has significantly affected this vegetation typology.

The biggest absolute extensions of altered areas in cerrados and other biomas occur in Mato Grosso, Tocantins and Pará, in this order. In Mato Grosso and Tocantins, the two states that have the greatest surfaces originally covered by cerrados, even greater than their forest surfaces, the extensions of altered areas of their cerrados are bigger than their converted forest areas to alternative land uses. Deforestation of cerrados in Rondônia and in Pará is also significant, but more in terms of the proportions of the original area of the bioma in these two states. Used areas x “abandoned” areas and main uses of land

According to estimates made by IMAZON11, and by Chomitz and Thomas8, on the basis of data and information of the IBGE agricultural census of 1996, it is possible to have a general idea, even if static, of the extension of used and non-used productive areas on the total of altered areas (that is, areas that had their vegetation cover removed/altered to give way to agriculture, silviculture or pastures) in the states of Legal Amazonia.

As per IBGE definitions, for each agricultural establishment the altered areas represent the sum of the areas of annual crops, perennial cultivations, planted pastures and planted forests, as well as other areas already altered and considered productive but non-utilized, and lands in rest. As can be seen, these areas amount to a total of 78.3 million ha in all of Legal Amazonia (Table 3.4). Still, by IBGE definition, the non-used productive areas are those altered areas that are capable of use for the establishment of crops/cultivations, pastures and plantations, but were not being used for such purposes, including lands not in use for a period greater than four years. The lands in rest are those habitually used for the establishment of temporary crops/cultivations, but were in rest for a period of less than four years, in relation to the last year of their establishment.

Table 3.4: Global extension of deforestation and of altered areas (converted to agricultural use; 000 ha) in Legal Amazonia – 1996



Area of forests


Area of cerrados and other biomas


Gross deforestation INPE/96





Global altered area IBGE/96


E = D - C

Altered area of Cerrado and other




































Mato Grosso
















































Source: (1) INPE – Monitoring of the Brazilian Amazon Forest 1998/99; (2) IBGE agricultural census 1996, obtained from IMAZON, 2000. Obs.: (-) inconsistent.

Note: Data on altered area of IBGE/96 refer to agricultural establishments. The definition of IBGE for altered area is the sum of areas of perennial and annual crops, temporary areas in rest, non-used productive areas, planted pastures and planted forests. It excludes, therefore, the areas of natural pastures, the intact cerrados and native forests. Given this definition, it does not seem too correct to qualify the non-used areas, totally, in the sense of degraded areas, and thereby in an “abandoned areas” condition.

Although the amount of non-used areas in Legal Amazonia reaches 16.5 million ha, which is almost 21 per cent of altered areas (Table 3.5) in the region - and one recognizes that low levels of land occupation, high rates of labour turnover, itinerancy and abandonment of arable areas are characteristics always singled out in analyses of the expansion of the agriculture frontier, and associated with disappointing results of colonization projects and settlements of rural population in remote areas of the Amazon - such conditions would be sufficient to allow us to qualify as “degraded” and “abandoned” all the areas classified as non-used, only if these areas were considered as intrinsically inapt for agriculture, of nil, or very low, agricultural vocation/aptitude.

In fact, those mentioned conditions also manifest themselves for other reasons, and not only because of “low agricultural aptitude” but also due to the: scarcity/lack of adequate technical assistance offered to the rural producer; changes in financing conditions and credit available to croppers/cultivators; occurrence of irregular climatic variations; and unforeseen price changes of inputs and agricultural commodities that alter investment plans for planting.

With these reservations in mind, and considering that the non-utilization of areas is a reflex of a set of factors, among which the natural vocation of lands versus the adaptation of crops is only one, from Table 3.5 it will be seen that Mato Grosso and Rondônia show the lowest indices of non-used areas, that is, these two states are those that have the highest rate of occupation of altered areas. It may be noted that Mato Grosso and Rondônia are also the two states of Legal Amazonia that had the highest annual agricultural GDP rates of growth in the 1970-1996 period, above the Amazon agricultural GDP growth as a whole, as depicted in Figure 3.2.

Table 3.5: Altered areas - used and non-used, or in rest, in Legal Amazonia –1996 (000 ha)



Global altered area IBGE/96


Non-used area


B / A



Area in use

C = A – B





















Mato Grosso






























Source: IBGE – agricultural census-1996, obtained from IMAZON, 2000

Note: According to IBGE, non-used productive areas include those that are good for the formation of crops, pastures and forests and that were not being used for such ends, including non-used lands, for a period of more than four years, and lands in rest, which are those habitually used for the planting of temporary crops, and that were in rest for a period of less than four years, in relation to the last year of their establishment.

Figure 3.2: Annual rates of growth of total GDP and of agricultural GDP of Legal Amazonia – 1970/1996

Figure 3.2 shows in which states of the region, for the 1970-96 period, the expansion of agriculture has been more intense and, therefore, where the activity has had a more influential role on total GDPs. The four states that had the highest annual growth rates of agricultural output were Rondônia (11.7 per cent), Mato Grosso (8 per cent), Pará (6.5 per cent) and Amapá (6.4 per cent). In terms of the whole of Legal Amazonia, the growth rate of agriculture and cattle-raising was 5 per cent average annual rate. in the period. In the states referred to, the agricultural GDPs grew faster than their total GDPs, which may have resulted in a significant deforestation impact due to the expansion of agriculture.

On the other hand, Maranhão and Amazonas are the states that have the highest percentages of non-used areas in relation to their altered areas. It may also be observed from the graph that Maranhão and Amazonas (besides Tocantins and Acré) are the states that, in the 1970-1996 period, had the highest growth rates of total GDP, with agricultural GDP growth rates relatively much lower, showing that their economies have expanded, more significantly, by stimuli from other productive activities than the agricultural ones.

Using the information of the World Bank study6, on the basis of the work of Chomitz and Thomas8, one can verify the variation of the use of land in the Amazon per rainfall zone. According to the data presented by the authors and included in Table 3.6, the major part of used areas of agricultural establishments (38 per cent) is found in the dry zone of the Amazon (which in greater proportion corresponds to the deforestation arch, and therefore where most of deforestation and conversion of areas to agricultural use occurs) and a small portion (3.2 per cent) in the humid zones of the region (zones that, comparatively, have a scarce population in the whole of the Amazon).

The dominant use of land in the altered areas is livestock breeding, with planted pastures representing 77 per cent of areas converted to agricultural economic use in Amazonia. Approximately 83 per cent of land in use in the dry zone is pasture, falling to 57 per cent in the humid areas (see Table 3.7).

As per Veríssimo, A., Arima, E. and Barreto, P.11, the present cattle herd (1996) is estimated at 32 million head of bovine cattle, and the respective total occupied area must be around 45 million ha (0.7 head/ha). Only in a few cases does cattle-raising reach an intensity greater than one head per hectare, in contrast with an average of 1.3 head per ha in the south and southeast of the country.

According to Andrade, E.7 while cattle-raising in the south, southeast and centre-south of the country is stabilizing, and in the northeast it has been falling since the 1992-93 drought, the north and centre-west regions have shown a surprising dynamism. In the last decade, the reduction of almost 5 million head of cattle in the northeast has been compensated by an equal or greater amount in the north. As IPAM data12 show, between 1992 and 1996 the bovine cattle stock has grown to more than 4 million head in the north.

Table 3.6: Use of land per rainfall zone in Legal Amazonia – 1996

Rainfall zone

Total area

(000 ha)

Area of agricultural establishments

(% of rainfall zone)

% of agricultural establishment area in agricultural use

















Source: World Bank

Table 3.7: Main uses of land in Legal Amazonia, per category of productive activity and per rainfall zone – 1996

Rainfall zone

Area in use


Area of pastures


Area of temporary crops (%)

Area of permanent crops (%)

Abandoned areas

































Source: World Bank; (1) includes planted forests and lands in rest.

Observation: Note that “abandoned areas” are included in ”area in use”, as per the quoted World Bank study. The authors did not define “abandoned areas”, nor why they included them in “agricultural use”. This inclusion differs, therefore, from the classification adopted in Table 4.

Cattle-raising, and its effects on Amazonian deforestation, have been thoroughly investigated in the search for explanations about progress and expansion of the agricultural frontier in the region. According to Margulis, S.13 there still is a lack of more consistent explanations on some variables and field observations, and one of the most important perhaps refers to the persistence of cattle-raising expansion in spite of various indicators that point in the direction of its low profitability. As Margulis points out, quoting Schneider, R. et al.6, until the end of the 1980s, various studies showed that cattle-raising had no satisfactory financial performance with the use of traditional technology (that is, it would only be feasible if there were fiscal incentives, subsidized credit, speculative gains on land and favourable input/cattle price relation). From the 1990s, some studies, like those of Mattos and Uhl14, started indicating positive feasibility for small dairy cattle production and cattle-raising in reformed pastures, and more recently Muchagata et al.15 confirmed low rates of return for traditional extensive cattle-breeding and better rates for small dairy cattle production close to roads in the region.

Notwithstanding these findings, according to Schneider, R. et al.6, the increase of bovine cattle-herds and of extensive cattle-raising of big and small animals continues without much economic and financial justification. In the past, economic incentives and stimuli granted by government (mainly subsidized credit and regional fiscal incentives) to the activity may have been an important factor in the maintenance and expansion of cattle-raising in Amazonia. However, nowadays this is no longer the case.

There are various suggested hypotheses that argue in favour of different factors that are not captured in the models which try to estimate “theoretical” return rates for the activity. For example, some of these factors linclude the following: net capital gains that are easily tradeable; land value increase and guarantee of land tenure by the existence of cattle; low risk of the activity; low levels of human capital and the reduced opportunity costs of labour, which are required by the activity and that are characteristics of the expansion frontier regions; smaller initial investments; ease of carrying out the activity; indirect benefits like animal traction; fertilizing; political and cultural power, etc.. Some of these factor were pointed out by Faminow et al.16 as advantages of cattle-raising in relation to other uses of land, and according to Schneider et al.6 still need more empirical verification.

Turning back to Table 3.7, and leaving cattle-raising aside, it will be seen that temporary and permanent crops occupy only 8.3 per cent of the total area in agricultural use in Amazonia, as shown by the 1996 agricultural census data compiled by Chomitz and Thomas8.

Although the Amazon, with its harvested area of temporary and permanent crops, participates with a still modest share (4.6 per cent in 1995) in the total harvested area of the country, this participation has increased considerably in the 1960-95 period: in 1960 it was only 1.5 per cent of the national total, the regional harvested area having jumped from 432 thousand ha in 1960 to 2.3 million ha in 1995. This is evidence of the significant increase in the participation of the Amazon harvested area in the total of the country. Until 1980 this share had only slightly evolved (it was then only 1.9 per cent of the national total). From the end of the 1980s, and especially from the mid-1990s, the share of the regional harvested area increased considerably in the Brazilian total (data from Monteiro Neto, A.10 ).

In the last 25-30 years the expansion of large-scale cultivation of grains in the Amazon – more specifically in the regions that constitute the so-called deforestation arch (the more meridional regions, drier and in contact with the cerrados of the central plateau) to the south of Maranhão, south/southeast of Pará, north of Tocantins and Mato Grosso – has attracted attention because of the surprising growth it has been showing. This expansion in grain production has also attracted international attention, not only for the environmental consequences of the extensive forms of production based on forest land conversion, but also for the perspectives of the country’s greater competitive potential in the international markets of agricultural commodities. The case of soybeans is illustrative of this picture. In fact, the most evident feature of grain production is the need for big land extensions to create the necessary scale for international competitiveness. On the other hand, if such an expansion occurs, without the adequate technological conditions, and in sites/areas that are also not adequate and do not provide economic and environmental sustainability for the crops, conditions for maintaining conquered shares in international markets and avoiding environmental losses and damages cannot be guaranteed.

According to data compiled by Monteiro Neto, A.10, considering the six main types of grains (rice, herbaceous cotton, beans, corn, soya and wheat), the Amazon evolves from a production of 204 thousand tons and a harvested area of 190 thousand ha in 1973, to almost 1.8 million tons and a harvested area of 1.27 million ha in 1997. This means that production has been multiplied by 8.8 times and harvested area by 6.7 times.

Despite the growth that occurred in these last three decades in the production and harvested area of grains, the Amazon’s participation in the country’s total grain production activity is still relatively small: in 1997, relative participation of regional production and harvested area in the national total were 2.4 per cent and 3.6 per cent, respectively, against 0.6 per cent and 0.7 per cent in 1973.

Although the region’s participation in the country’s total production is still small, data show that the importance of the Amazon for growing grains has been increasing due to the: (i) continued reduction, since 1988, of the total harvested area of grains in the rest of the country, meaning that the expansion of grains in the region has been compensating retraction in the other areas of Brazil; and (ii) as well as to the extensive expansion of the activity (that is, the increment in the relation between harvested area in the north to that of Brazil as a whole), which has been superior to the increment of the participation of the region’s production of grains in the national total, in the last three decades, revealing that growth is taking place more on account of the expansion of areas than to the productivity gains.

Finally, in the face of this expansion in grain cultivation in Amazonia, one may enquire about possible competition, or substitution, between cattle-raising and other large-scale agricultural crops in the region. As pointed out by Margulis, S.13, “there have been indications that large-scale agricultural production in the region does not seem possible without a previous ‘adaptation’ of soils with cattle-breeding, particularly in the case of highly technically demanding crops like soya beans. Not for other reasons, perhaps, do soya crops only exist in the cerrado areas of Mato Grosso, Tocantins and Maranhão; there still is a lack of technological control and dominance in soya beans, before it outweighs cattle-raising in the south of Pará”. The “deforestation arch”

Brazil is one of a few developing countries that have a modern and comprehensive forest cover monitoring programme for most of its territory. Known as PRODES (Project of Brazilian Amazon Deforestation Estimates), and coordinated by the National Institute of Spatial Research (INPE), the monitoring of the Amazon forests has been systematically carried out for many years (since 1988) through the employment of satellite remote sensoring techniques, using the images of the north American Landsat satellite, in coloured compositions, in the scale 1:250,000, which provides the identification of deforestation areas bigger than 6.25 ha.

As Krug, T.17 observes, the singularity of this programme is associated with the frequency with which deforestation sites in the Brazilian Amazon are verified, as well as with the geographical scope and extension of the area in question. In most of the countries (mainly the tropical ones), their forests are monitored every five or 10 years, and in a way that is not necessarily comprehensive. Therefore, Brazil has been generating annual estimates of the mean average rate and extension of gross deforestation since 1988 for the entire Legal Amazon region based on an annual basis of 229 images of the TM-Landsat satellite. Each year, an overlay containing all deforestation sites that were identified until last year is placed over each image of the Amazon. Polygons are then delimited if they present a “clear cut” pattern of deforestation. The fact that PRODES is carried out every year, involving alterations of forest areas that promote soil exposure, makes identification of an affected area by deforestation unequivocal, without the need of a field verification.

Additionally, this methodology allows the crossing of deforested areas with other types of data, such as the municipal limits, the vegetation map of Brazil (Radambrasil surveys, IBGE) and others. The significant progress provided by the continuous development of the technology of geographical information systems has also permitted PRODES to evolve from a simple presentation of illustrative tables and deforestation graphs to generating relevant data to public, federal or state institutions of environmental control and inspection, mainly for the purposes of carrying out prognoses, in which knowledge of the geographical location and spacing of deforestation over time is fundamental.

Deforestation, in the monitoring implemented by INPE, is understood as the conversion of areas of primary forest physiognomy by human actions for the establishment and development of agro-silvo-pastoral activities. Deforestation detected by PRODES is a “gross deforestation”, that is, only altered areas by anthropogenic actions of forest cover removal known as “clear cut”. So, the net result between deforested areas and regrown areas (succession or secondary regeneration) is not taken into account, i.e. one does not consider areas that are already recuperated or in process of recuperation. Forest cover alterations due to exploitation of selective timber cutting activities that do not immediately result in the total removal (clear cut) of forest cover, are also not captured. Areas affected by forest fires are not included either, though these are the object of other specific INPE reports (since 1988 Brazil has had a system of national prevention and combat of forest fires (PREVFOGO) under the coordination of IBAMA. In May 1998, a national programme for the prevention and control of burnings and forest fires in the deforestation arch (PROARCO) was created, and has since been coordinated by IBAMA, involving federal institutions like INPE and other state institutions – this programme is based on a monitoring system of daily maps of heat foci, fire risk maps and alert systems for the municipalities of the Amazon, all stemming from daily images of the NOAA and GOES satellites.

The last report of INPE18 (May 2000) shows the complete and final estimates of deforestation in Legal Amazonia for the period 1998/1999, and the provisional estimates for total deforestation that occurred in the region in 1999/2000. Tables 3.8, 3.9 and 3.10 present the deforestation data per state of the Amazon.

Table 3.8: Extension of gross deforestation (km2) from January 1978 to August 1999

Amazon States

































































Mato Grosso

































































Brazilian Amazon













Source: INPE, 1999

The utilization of GIS by PRODES/INPE, with the crossing of the deforestation polygons and the vegetation map of IBGE, has made it possible, for example, to verify that around 63 per cent of identified deforestation in the last five years (average percentage from 1995 to 1999) has occurred in areas of dense and open rain forests. It has also been possible to build classes of deforestation sizes that allowed the identification of deforestation patterns for each type of affected vegetation (pioneer formations, forested campinarana, cerradão, contact/transition region, seasonal forest, open and dense rain forests).

According to Krug, T.17, the continuous monitoring of the forest cover of Amazonia and the crossing of information by PRODES permit verifying that the major part of deforestation activities in the Amazon (approximately 75 per cent) has traditionally concentrated in a limited and specific portion of the Amazonian region, that has come to be known as the “deforestation arch”. The arch covers part of the states of Acré, Rondônia, Mato Grosso and Pará, and is covered by 50 TM-Landsat images, that is, 75 per cent of total deforestation is concentrated in approximately 20 per cent of the images that cover the whole region (see Map 3.8).

Table 3.9: Average rate of gross deforestation (km2/year) from January 1978 to August 1999

Amazon States




























































Mato Grosso




























































Brazilian Amazon












Source: INPE, 1999

Compounding this deforestation arch, the southern areas of Legal Amazonia, drier and in contact with the central plateau cerrado regions, due to the lower cost of land, the greater availability of access roads, the greater relative proximity of bigger urban centres of the rest of the country, the cheap supply of labour, the easier clearing of forest areas with less vegetation density – areas that correspond to the regions of the south of Maranhão, south/southeast of Pará, north of Tocantins, and Mato Grosso – have in the last decades attracted the attention of corporate business groups that have shown interest in large-scale grain cultivation, as well as in the expansion of cattle-raising and timber exploitation associated with the wide availability of wood that comes from deforestation.

Table 3.10: Average rate of gross deforestation (%/year) from January 1978 to August 1999

Amazon States




























































Mato Grosso




























































Brazilian Amazon












Source: INPE

Map 3.8: The deforestation arch in Legal Amazonia

In this strip of land, the width of which varies between 300 and 500 kms, the public power has been intensifying its investments in infrastructure, expanding and improving the transportation routes, increasing energy generation capacity, etc. All of this region, which has shown a greater dynamism and greater supply of jobs, has also been a place of intense social mobilization, with an overloaded infrastructure of public social service, where poor supply of basic needs has been provoking an increase in social tensions.

In the more oriental portion of this “arch”, a wide institutional public and private “consortium” has determined a large polygon (known as the area of the “Great Carajás” programme, based on the Carajás-Iron Mining Project) where the development of strategic actions would benefit from the modern infrastructure of the Carajás/São Luís railway, federal and state roads, the Itaqui port in Maranhão, the Araguaia/Tocantins waterway and other available large physical investments aimed at ensuring the bases for modern technology agri-businesses. In the more occidental extremity of the “arch”, another similar “consortium” has provided feasibility to one of the most important multi-modal routes of regional channelling of production: the waterway of the Madeira/Amazonas rivers, placing wide agricultural areas located in central Brazil in an advantageous market competitive position. Map 3.9 provides a view of the main routes of penetration and access to the regions of the “deforestation arch”, and others, in the Brazilian Amazon.

The crossing of PRODES satellite images information with that of the vegetation map of Brazil (IBGE) and the employment of GIS techniques has provided a means of extracting some conclusions about the profile of deforestation in the Amazon, particularly in the “deforestation arch”. Thus, for example:

• in the 1995/99 period, in which it was identified that 63 per cent of deforestation occurred in areas of dense and open rain forests, it was also found that around 20 per cent of the total number of deforestation polygons in these forest areas are smaller than 50 ha, and approximately 10 per cent are larger than 1,000 ha (average percentages for the 1995/99 period), while a major part of deforested areas in the contact/transition regions (around 21 per cent) are bigger than 1,000 ha. These differences in the pattern of deforestation serve as indications or empirical evidence for the comprehension and analysis of the possible causes or factors of the deforestation process. In the examples cited, one may speculate, for instance, that big extensions of deforested areas in regions of transition/contact may be linked to the expansion of large-scale crops or cattle-raising, while the small deforested areas would be linked to small-scale subsistence agriculture or occupation of lands for the purpose of agrarian reform;

Map 3.9: Legal Amazonia – national integration and development axes of the “Avança Brasil” programme

• in relation to the whole period of 1978/99, a significant variation of the gross deforestation rate has been observed each year, having ranged from around 29,000 km2 (1994/95, the largest observed rate since 1978) to 13,000 km2 (1996/97, second smallest rate estimated since 1978), a difference that occurs in only two years (from 1994/95 to 1996/97). As pointed out by Krug, T. op. cit., much debate and speculation exists about the possible causes that may explain these variations, mainly those of an economic nature, such as the changes in the economic and monetary stabilization policy of the country with the “Real Plan” in 1994, decrease in the real value of lands in the Amazon, etc. Nevertheless, one still needs a more comprehensive explanatory model, capable of justifying the variations that occurred in the deforestation rate in the Amazon and that can provide more secure prognoses of the behaviour of deforestation;

• another type of information provided by the surveys of INPE/PRODES refers to the quantification of primary forest areas that are affected by activities of selective timber exploitation type, in sites that are not, or will not be, immediately subject or suitable for deforestation or clear-cutting, in the process of converting land to agricultural use. This has been a point of controversy in the literature and in the analyses about the “true extension” of deforestation in the Amazon (as examples of studies that criticize data/information of PRODES/INPE, and that try to show that the extension of deforestation in the region is larger than that pointed out by INPE, one may quote the work of Nepstad, D. et al.19, and as reply the work of Krug, T. et al..20). As per the specific surveys carried out by INPE, from TM-Landsat images that showed a pattern associated with the selective type of timber exploitation (in which only a few high value species per unit of area are extracted in the forests) this leaves a still significant area covered by forests. The analysis – done for each year, from 1988 to 1998, and based on visual interpretation – identified 26 scenes (among the 229 that cover the whole of Legal Amazonia) with that pattern, with different intensities of exploitation. The data obtained by the analysis provided the following conclusions: (a) the average annual increment (extension) of areas associated with selective cut were only of the order of 2,000 km2; (b) of the delineated polygons for the 1988/98 period, 40 per cent ceased to show a spectral pattern of selective cut type, which suggests that the affected areas were in process of recuperation, i.e. the pattern of these areas began to be confused with that of primary forests; (c) 15 per cent, therefore, of the delineated polygons, in the period, began to have a pattern corresponding to clear-cut activities of effective conversion to agricultural use; (d) the remaining polygons still presented, in 1998, a pattern of selective cut; (e) the average time for an area submitted to selective cut to stop showing signs of this type of activity in the images is close to four years.

Another interesting description of the evolution and profile of deforestation in the Amazon, that has already been used for optimizing “command and control” actions of environmental monitoring, inspection and policing in the region, either by IBAMA or by OEMAs, is that which refers to the spatial location of deforestation by municipalities that mostly concentrate these forest-cover conversions to other land use. Map 3.10 is the result of joint work carried out by the Secretariat of Amazonia/Ministry of the Environment (SCA/MMA) and IBAMA, and shows that the great majority of municipalities that mostly account for deforestation in Legal Amazonia lie in the “deforestation arch”.

According to data compiled by Menezes, M.21, as per Table 3.11, 47 municipalities responsible for 50 per cent of deforestation that occurred in the states of Mato Grosso, Pará and Rondônia (the three states that contribute with 86 per cent of total deforested area in the region) cover an approximate surface of 568,000 km2 (or 13.6 per cent of the whole surface of the region). If the 139 municipalities that account for 90 per cent of deforestation in these three states are taken into account (which amounts to 77.4 per cent of deforestation in Legal Amazonia as a whole) one comes to a total surface of 1,238,000 km2 (which is almost 30 per cent of the region’s surface).

This finding, although showing that there is a concentration of deforestation in a gigantic extension of lands, but restricted to a relatively small number of municipalities, is more important for the strategic content of its political message or implication for the control of deforestation in the sense that the responsibility for monitoring, control and public decision-taking concerning the expansion (or reduction) of deforestation should also fall on the shoulders of the local public power (municipalities) and not only on federal or state fields of action. But such implication also poses a further problem: local public decision-taking on issues like deforestation tend to be very much influenced by local vested interests (interests of people directly involved in and willing to expand conversion of forests to other uses) and by short-term financial gains that local public power see as forthcoming by the expansion of whatever production can be created in the municipality. Moreover, these large municipalities, whose territories are bigger than many countries in the world, are indeed very poor and certainly do not have the means to devote adequate efforts to controlling and combating deforestation.

Map 3.10: Legal Amazonia – distribution of municipalities that mostly concentrate deforestation – 1997

Table 3.11: Number of municipalities that contribute with progressive deforestation rates (50 to 90 per cent) in the states of Mato Grosso, Pará and Rondônia, with their respective shares in regional deforestation and percentage classification in relation to all 624 municipalities of Amazonia


Number of municipalities

Share (%) of municipalities in region’s deforestation

% of Municipalities in the region’s total (624)

Total area of municipalities (km2)

% of total region’s surface































Source: Menezes, M.

This points to the need for establishing institutional partnerships, governmental and non-governmental (such as, for example, the implementation of federative pacts among the Union, states and municipalities for the execution of environmental policies and decentralization of environmental management, etc.) with the purpose of optimizing, and making more effective, the establishment and operation of an integrated system of environmental monitoring and control that makes use of modern instruments of remote sensoring, licensing and inspection, capable of operating in such a vast region like Amazonia. Private areas x public areas and conservation units

Private areas

According to the latest statistics from the “Registry of Rural Properties of Brazil” (1998), produced by INCRA, the structure or profile of private land tenure in Legal Amazonia can be depicted as in Table 3.12. The data from this Table reveal an extremely concentrated pattern of private land distribution in the Amazon region, a concentration much stronger than that which prevails for Brazil as a whole, according to INCRA data. As will can be seen from Table 3.12, almost 90 per cent of all rural properties in Amazonia hold less than 20 per cent of the total area occupied by these estates or, in other words, more than 80 per cent of the total area of rural properties are concentrated in a bit more than 10 per cent of these properties.

Note: The rural modules are defined in Law 4504/64 (Land Statute) as the necessary area to guarantee the economic subsistence of a cropper and his family; the dimensions of the modules of rural properties are established by municipality of the Amazon, whose maximum dimension, per module, is the following: in Acré, up to 100 ha; in Rondônia, up to 60 ha; in Amazonas, up to 100 ha; in Roraima, up to 100 ha; in Pará, up to 75 ha; in Amapá, up to 70 ha; in Tocantins, up to 80 ha; in Mato Grosso, up to 100 ha; and in Maranhão, up to 75 ha.

Table 3.12: Profile of private land tenure in Legal Amazonia – number of private properties (estates) and total area, per classes of area size of the properties – 1998

Classes of area size of properties (ha)

Total number of properties

Total area of properties

In units

In percentage

In 000ha

In percentage





Up to<25 ha







25 to<50 ha







50 to<100 ha







100 to<200 ha







200 to<500 ha







500 ha to more














Up to ≤ 1 module







>1 to ≤ 2 modules







Source: Estatisticas Cadastrais – INCRA – 1998.

From the point of view of taxation on land property, and corresponding to what is established in Law 9393/96 that instituted the new ITR (and in compliance with the dispositions of art. 153, §4, of the Brazilian Constitution, that the ITR cannot be levied on small rural properties when these are exploited by the owner alone, or with his family, provided he does not own another property), small rural properties or estates are exempt from the tax and, in the case of Amazonia, are defined as properties with area equal to or smaller than 100 ha if located in a municipality of western Amazonia, or with area equal to or smaller than 50 ha if located in a municipality of eastern Amazonia.

Based on these definitions, and using the data from Table 3.12, one can estimate that at the very least there would be 219,873 small properties in Legal Amazonia (56.4 per cent of the total of registered properties in the region) that would hold around 14.5 million ha (only 8 per cent of the total area of properties) of land.

It is interesting to compare this information about the structure of private property of land in the Amazon region with some of the findings of PRODES, as per the INPE surveys. As we have remarked (under, it has been identified that from 1995 to 1999 around 20 per cent of the total number of deforested sites in areas of dense and open tropical rain forests were smaller than 50 ha, and that it had been speculated that these small or smaller deforestation sites could be associated with small-scale or subsistence agricultural activities, or occupation of lands by settlements of agrarian reform, that is, that deforestation in the small properties of Amazonia would also have a relevant role in the composition of the global deforestation rate in the region, despite the weight of bigger deforested sites in the properties with bigger areas.

Surely, if we separate the very small properties, with less than 25 ha (see Table 3.12), and the total area that they occupy, the impact or contribution of the deforestation attributed to them on total regional deforestation should be insignificant. But, as we move up on the scale of the sizes of rural property areas in the region, and consider that, for the Amazon, small property is something that goes up to 100 ha (inclusive), then we can see that the group of small properties may indeed have a more relevant weight on the global regional rate of deforestation, without necessarily transforming them into the main deforestation agents in the region.

The implications of this finding for the purposes of control, monitoring and policing of deforestation, and for the enforcement of the legislation that regulates it, are important. One cannot underestimate the weight of these small properties, and pretend, for example, to concentrate actions only in the monitoring and control of big properties (something that would be less costly and operationally easier given the possibility of using modern technologies to identify more quickly the big deforestation sites and to arrive sooner at the places with policing and inspection teams). Much less should one concede a more favourable or acquiescent treatment to “average” size properties (up to 400 ha, for example, as has been the case of Normative Instruction 03, dated 04.03.2002, of the Ministry of the Environment, still in force, that attenuated requirements for the authorization of deforestation in properties up to such an area) on the assumption that a more rigid and systematic control would be necessary, in fact, only for bigger properties (more than 400 ha) since these would be the ones that would generate a significant impact on the global rate of deforestation. As can be seen from Table 3.12, the extension of the area occupied by properties whose individual size reaches 400 ha, in the entire Amazon region, is quite significant and has even more weight in the total. It is a concrete possibility, therefore, that deforestation in these properties has a more than negligible influence on global regional deforestation.

Another important aspect of the deforestation question, be it in small or big properties, is that related to the supply to the markets of native timbers that become available and originate from deforestation that occurs in these properties. Although the relative weight of “small”, in terms of deforested area, may be less than that of “big” deforested areas, this proportionality, or contribution, may be different, or less apparent, in terms of the global supply of timber to the wood industry of the region. This is a more complex issue (with which we shall deal further below) with serious implications for the systems of enforcement of environmental norms and regulations in the control and monitoring of deforestation and of associated activities, including those of timber exploitation, particularly with regard to ensuring an authenticity/legality of the origin of native timber (derived from authorized deforested areas, small or big) that enters the country’s markets.

Within the structure of private property of lands in Amazonia, one should highlight the official colonization and agrarian reform settlement projects existing in the region. Until August 2000, according to INCRA data, and as depicted in Table 3.13, there were 1,460 rural settlement projects in Amazonia, involving more than 300,000 officially settled families and a global area of more than 17 million ha (as per INCRA statistics, outside Amazonia – that is, in the rest of Brazil – 2,482 agrarian reform settlement projects had been established, involving 177,000 families and a global area of almost 4.9 million ha, which corresponds to an average estate of 27.7 ha per settled family).

As for the official colonization projects created in Legal Amazonia, up to August 2000 (in the recent past the so-called “private colonization” was an option largely used in the region, particularly in states like Mato Grosso, for instance) the data indicate that around 13.3 million ha have been involved in this modality of transference and occupation of lands in the region (while in the rest of Brazil only 138,000 ha would have been the object of colonization projects). In general, in Amazonia, the average area of rural estate per family involved in a colonization project has been bigger than that of the agrarian reform settlement projects. Based on these data one can see that up to August 2000 at least 30 million ha of land had been utilized for and transferred to peasants and rural producers in settlement and colonization projects in Legal Amazonia.

Table 3.13: Agrarian reform settlement projects created in Legal Amazonia – total area, number of projects and settled families, per state, up to August 2000


Area (000ha)

No. of projects

No. of families


(in ha)











Mato Grosso








































Source: INCRA, in MMA

Note: It should be noted that the data on registered properties of the “Estatisticas Cadastrais” of INCRA (of Table 3.12 above) include the properties that are part of settlement and colonization projects. Therefore, the figures of the present Table would be included in the data of Table 3.12, except for the amount that would refer to the period after 1998.

In general, and on the basis of observations in the region, it is common to hear (cf., for example, Fearnside, P.22 and Sawyer, D.4) that on average “small rural producers or family croppers can only deforest up to 2-3 ha/year, with family labour, this being a pattern that is reflected in the deforestation behaviour in areas of rural settlement”. Only for comparative purposes, and without the intention of giving the impression that small rural producers could be the main deforestation agents, taking as a basis INCRA’s number of officially settled families in Amazonia (at least 300,000), as well as the average of 2.5 ha/year of forests cleared per family/cropper, it may be assumed that a total of around 700,000 ha per year of deforestation could be attributed to this part of the population in Amazonia. This would be a conservative figure, since, as Sawyer4 well observes, in a rural population of 8 million people in Amazonia, or almost 2 million families, it is reasonable to assume that there are at least 1 million family croppers/producers in the region. The number of families and people involved in spontaneous settlements (non-official) in Amazonia - in adjacent areas to colonization and settlement projects, as well as in other new areas (where families are camped and will be incorporated into new settlements) or in areas of old frontiers - must amount to a number that is at least equal to what has been officially settled.

As is notorious, the Brazilian agrarian policy has for many years given emphasis to the occupation of the Amazon. In the 1970-1984 period, 71 per cent of discriminated lands in the country were in this region, with much less accomplishment in the other regions of the country. The insufficiency of other social policies would add to this, increasing the incentives to the spontaneous migration process directed to the Amazon. The region still continues to participate in the solution of problems of the other regions of the country, despite current intra-regional predominating over inter-regional migration, as we remarked before.

According to INCRA, and based on Weiss, J.23, the proportion of beneficiary families in settlements of Amazonia, in relation to the total of settled families in Brazil as a whole, has been undergoing a process of reduction over time. In 2001, the regional target of INCRA for settling families in Amazonia corresponded to only 27 per cent of the national total, and in 2002 this proportion was reduced to 14 per cent. Another indication of this trend can be obtained from the relation of settled family/camped family: while in 2001 around 2.5 families for each camped family had been settled in Amazonia – 20,310 settled families in 2001 to 7,982 families already camped in 2002 – the target for 2002 has been reduced to 1.05 settled family/camped family in the region (this is more in line with the determinations of Law 4504/64 – the Land Statute – whose orientation commended agrarian reform for families in their areas of origin).

With this type of change in the implementation of the agrarian reform policy by INCRA, the regional target for the Amazon is reduced from 16,348 to 8,400 families to be settled. If such a trend is maintained, one could perhaps foresee a future reduction of the environmental impact of the agrarian reform policy in the region. Nevertheless, as stressed by Weiss, J., op.cit., the performance of the regional superintendencies of INCRA continues mostly to be measured in terms of the number of settled families against the established annual targets. Only more recently have “sustainability indicators” come to be considered as parameters for the evaluation of that performance.

Another area of agrarian policy in which the actions of the government have more recently come to be implemented in a more effective way is that of land regularization. It is interesting that this be mentioned here, because seeking land tenure regularity has a direct affect on the registry of lands in Brazil, and in particular in Amazonia, and thereby the data and information of this registry.

In order to bring the occupation of lands in the Amazon to a more normal pattern, and to know the holders of large areas, whether or not occupied by private agents, INCRA has been adopting different measures to combat irregular latifundia and irregular occupation of lands (such irregularity happens, in general, through the illegal way in which lands end up in the hands of private initiative, either under good faith acquisitions or under criminal methods that forge false documents (methods known as “grilagem”).

From 1999 to 2001 INCRA has strongly attacked the holders of rural properties, using legal requirements to know better the reality of land structure in the Amazon, on the basis of the Land Statute, the Brazilian Civil Code, the Forest Code and the Law of Public Registries.

According to information contained in Weiss, J., op.cit., INCRA has detected more than 100 million ha of land in the whole country, of which 70.5 million ha in the Amazon, that have been identified as irregular lands (“grilagem”). Table 3.14 presents the figures, per state of Legal Amazonia, of private irregular

Table 3.14: Areas identified as irregular (“griladas”) in Legal Amazonia – 2000


Million ha









Mato Grosso












Source: Ministry of Agrarian Development, in Weiss, J.

lands. In a partnership with the Secretariat of Federal Income (SRF), and on the basis of Law 10267/01 that is the basis for the National Registry of Rural Properties, INCRA has rescinded the records of individual properties larger than 10,000 ha (around 3,000) and of individual properties with areas between 5,000 and 10,000 ha (around 600). This, undoubtedly, reveals the fragility and deficiency of the registry statistics of INCRA for 1998 (and other years), last published data of the institute, and that have been used here to describe the agrarian structure of the Amazon. The data do not reflect a very trustworthy reality of the profile of private land distribution in the region. The elimination of records of properties taken as irregular before INCRA and SRF creates a loss of identity for these estates, which results in a series of obstacles and restrictions that prevent these properties (and their alleged owners) from undertaking any sort of businesses, firming contracts, etc.

An important aspect of Law 10267/01 is that it foresees the national registry of rural properties linked to other registries of federal institutions, like SRF for example, and includes requirements such as the geo-referencing of the vertices of the properties’ perimeters, particularly for the transference of properties. The effective implementation of the Law will solve a good part of the existing conflicts in relation to rural properties in the Amazon. Their holders, for instance, will be liable to be held responsible, in a more effective way, for environmental damages committed on their properties, damages that will come to be captured through satellite images. Undoubtedly, this is a particularly important point for the conception and building of a control and monitoring system for public environmental agencies (at federal or state levels), the attributes and competence of which are concerned with the enforcement of environmental legislation relative to the adequate use of land in the country (we shall comment more on some of these control systems that are already being used, at state level, in the Amazon region).

Public areas

Although a good part of the land in Amazonia is already in private hands – as per the 1998 statistics of INCRA, the total would be 181.16 million ha according to Table 3.12, or 110.66 million ha if we deduct the 70.5 million ha considered irregular by INCRA, cf. Table 3.14 – the major part is still in the public domain. Table 3.15 summarizes the general picture of land property in Legal Amazonia, highlighting the total of private areas and the areas in the hands of public power, the latter divided into public areas already legally defined (indigenous reserves, conservation units of strict protection and conservation units of sustainable use) and public areas still without legal discrimination or determination of destination or use.

As will be seen from Table 3.15, there are in Amazonia around 103 million ha legally defined as indigenous reserves (in the last 5-6 years, federal government demarcated more than 22 million ha of indigenous reserves in the region), 13.9 million ha determined as strict protection conservation units (national and state parks, biological reserves, ecological stations, etc. whose target established by federal government is to increase the area of this category of conservation unit to 10 per cent of the total area of the region, that is, to create more than 37 million ha of strict protection units in Amazonia) and 34.2 million ha defined as conservation units of sustainable use (national forests, extractivist reserves, reserves of sustainable use, and others). The states of the region with the largest surface areas legally defined as public areas are: Roraima (62.7 per cent) and Rondônia (43.3 per cent). Mato Grosso (16 per cent) and Tocantins (16 per cent) are areas that have the smallest proportions of identified public areas in relation to their territories.

There is at least 38 per cent (193 million ha) of land in Legal Amazonia in the hands of the public power, federal and state, that would not yet have a defined destination (areas in the power of INCRA, but still not destined for settlements, are part of this total). Although this represents an enormous quantity of land in the entire region, the situation is not so uniform at the state level. Some states, like Tocantins, Roraima and Rondônia, have smaller portions of undefined public areas than the average for the whole of the region (Mato Grosso would be the state with the smallest share, but given the extensive area of private land registries considered irregular by INCRA – 22.8 million ha – a good quantity of these lands could be reverted to public domain).

As for the quantity of area still in the hands of the public power (at least 193 million ha) there should be no lack of land for accomplishing the 10 per cent target of strict protection areas in Amazonia, as well as the target of the Brazilian national forest programme of creating a network of national forests in the same region with at least 50 million ha (there are already around 17 million ha of national forests in the Amazon).

The conflicts between land privatization policies in Amazonia and those of creation and destination of public lands for conservation units (be it for strict protection or for sustainable use) are a registered mark of the national geo-politics for the region and of its process of occupation. The creation by federal government of a federal conservation unit in a state of Amazonia often displeases the state government (even if lands did not belong to the state government but to federal government) who would prefer, for example, to see those lands occupied by agricultural activities. Most of the time, the process of choice itself, and of definition and creation of a conservation unit, whether for strict protection or for sustainable use, has been marked by extreme positions of environmentalists who only accept strict protection for environmental conservation areas in Amazonia, and thus try to create difficulties when it is a case of creating areas for sustainable use in the region.

Nevertheless, in the last decade, there has been progress in the creation and establishment of conservation units in the region. INCRA, itself, by determination of the government and of the Ministry of Agrarian Development, is defining and transferring to IBAMA a total of 20 million ha of lands in the Amazon (resulting from areas regained by the policy of land regularization and rescinding records of properties considered irregular) for the creation of conservation units, out of which 14 million ha will be destined for the creation of national forests. The creation of conservation units in the Amazon relies today on a strong apparatus of technical and zoning studies that serve as a basis for the selection process and choice of the areas. In this respect, the studies promoted by the Ministry of the Environment should be mentioned: (i) the project of studies for the definition of priority areas for conservation of bio-diversity of the Amazon bioma, which was summed up in the Macapá Consultation Workshop in September 1999; and (ii) the study for the identification of areas with potential for the creation of national forests in Legal Amazonia. Maps 3.11 and 3.12 show the potential and priorities for the creation of these conservation units.

Map 3.11: Priority areas for the conservation of bio-diversity in Amazonia

Source: Macapá Consultation Workshop, September 1999

Map 3.12: Potential and priority areas for national forests in Legal Amazonia

There are various criticisms of the national system of conservation units of Brazil (Law 9985/00 established and defined this national system), among which those that allege that the government has been incapable of satisfactorily implementing this system, or that these units have only been created “on paper” and are not in fact protected (especially the units created in Amazonia), due to a lack of adequate infrastructure, personnel under-staffing and insufficient action by IBAMA to keep them. No doubt, the Brazilian Government and IBAMA are in need of means to better implement the system of conservation units in Amazonia.

But the simple organization or zoning of the territorial space in the region through the mere transformation of lands in areas of public domain by the creation of protected areas, such as indigenous reserves and conservation units – not to mention their benefits for the protection of bio-diversity and protection and guarantee of indigenous populations – has an extremely positive effect in safeguarding these spaces from the expansion of deforestation and privatization of lands through deforestation. And this has been a fact, as rightly demonstrated by Arima, E.24, of superimposing the conservation units and indigenous reserves map on the map of heat foci in Amazonia identified by NOAA satellite. As will be seen from Maps 3.13 and 3.14, it becomes clear why there is no fire in enormous areas of the region. The occupation, deforestation and burnings are inhibited in these vast areas, even without the presence of IBAMA’s inspection and policing, because these are public power lands and this helps to make invasion, occupation and resource depletion more difficult or infeasible.

To conclude this section, a few points that serve as evidence of how the irregularity and insecurity of land tenure in the Amazon are associated with deforestation in the region should still be noted. An irregular and unclear land tenure framework is a characteristic feature of many regions in Brazil, particularly in areas with vast forest cover, as in Amazonia, but even in some areas still covered by the Atlantic forest.

According to agrarian and civil legislation in Brazil, people can claim land in the Amazon (as in other regions of the country) provided they have occupied a certain area for a certain period of time (a person who does not own lands can acquire up to 50 ha of rural land, making it productive, after an uncontested possession of five years) and that they have “improved” the area or “made it productive”. Historically, to fell trees and to “clean” the land for crops and pastures have been considered, by INCRA, as an “improvement” or a “benefit”. Deforestation has therefore been a rational decision for those that seek to obtain the possession and property of rural land. In the process of occupation of lands, deforestation ends up being a means for the establishment of private ownership of lands, a pre-condition for the legalization of the property.

Land is incorporated into the private sector through various legal (leasings, public biddings, concessions, agrarian reform settlements in accordance with the Land Statute/Law 4504, usage based on occupation and possession, etc.) or illegal mechanisms (“grilagem”, false registrations in notaries public, etc.). The possession of land, in general, is proved by lasting occupation and by deforestation. The irregular or illegal forms of getting possession or acquisition of land in Amazonia have been constant in the occupation process of the region, which contributes to increasing the inequality of access to land and use of natural resources, besides provoking serious conflicts and violence in the struggle for the possession of land.

Map 3.13: Heat foci – NOAA – June to December 1999

Map 3.14: Federal conservation units, indigenous reserves and heat foci in Amazonia

In Amazonia, the irregular and insecure regimes of land tenure have been indicated as one of the main factors that favour deforestation and serve as dis-incentives to SFM. Various authors, such as Weiss, J.23, Margulis, S.13, Viana, V.25, Prado, A.C.26 and others, single out the fact that the irregularity and insecurity of land ownership are an incentive to degradation of forest resources, favouring an attitude of converting gains from the exploitation of these and other natural resources, with quicker liquidation and loss of natural capital. When possession/property rights are not well defined, or are insecure, the planning horizon of involved agents reduces enormously, so that losses with the liquidation of natural capital do not get incorporated into their short-run decisions. The irregular and insecure tenure of land, therefore, favours deforestation and dis-favours environmental conservation and economy in the use, or sustainable use, of the natural forest resource. In this sense, therefore, such a framework of land tenure is something that militates against SFM.

Far from being a consensus, this issue is much disputed in the literature. So, for instance, while the team of Brazilians that participated in the elaboration of a World Bank sponsored study, “Forests in the Balance: Challenges of Conservation with Development – An evaluation of Brazil’s Forest Development and World Bank Assistance”27, is categorical in stating that clear and secure land possession and property rights are a prerequisite/condition for the conservation of forests in the Amazon through SFM (without which the chances for forest management to compete with the alternative activities that remove forests remain poor), the other authors of the document consider that there is no certainty, or consensus, that a framework of more security and regularity of land tenure contributes to reducing deforestation.

According to the cited document, and to the arguments of the authors referred to, on the contrary, “to accelerate the formal processes of ensuring clearer and more formal land tenure will probably increase the deforestation rate in small properties, since these will have easier access to credit”, or that, “because greater security in land tenure leads to more investment in land, this could equally well apply to investments in forest management or reforestation as to investments in deforestation/removal of forests for the establishment of agriculture or cattle-raising”.

Authors like Weiss, J.23, for example, are even more categorical in stating that a situation of secure and regular possession and ownership of lands in Amazonia is a sine qua non for better control of deforestation and for reducing the dis-incentives to SFM for which the current unclear pattern of land tenure is responsible. As the referred author says, “more important than the illegality of forest exploitation (“illegal logging”) in the control of deforestation and in providing dis-incentives to forest management is the illegality of possession/property of lands in the Amazon”.

Table 3.15: Use of land in Legal Amazonia – private areas x public areas and conservation units




(000 ha)


Area of private properties


(000 ha)



Public Areas (F) = C + D + E




(A – B – F) / A


Indígenous reserves

(000 ha)




CUs Integral protection

(000 ha)




CUs sust. use

(000 ha)




(000 ha)





















































Mato Grosso














































































Sources: MMA, “Causas e dinâmica do desmatamento na Amazônia”

3.4 Dynamics of timber exploitation and timber industry in Legal Amazonia

3.4.1 The numbers in the timber industry of Amazonia

Perhaps the best and clearest way of presenting an overview of the profile and dynamics of timber exploitation activity (industrial timber exploitation and processing) in Amazonia is to use the evidence provided by some recent research work such as that of Schneider et al.6, IMAZON9, Veríssimo, A. et al.28, Veríssimo et al.29, Uhl, C. et al.30, Nepstad, D. et al.19 and Scholz, I31. These authors, after some years of research and assessment of the deforestation process under way in Legal Amazonia, have produced excellent syntheses of “the production mode” of timber exploitation and timber industry activity in the region. According to them, Legal Amazonia would nowadays produce approximately 90 per cent of the industrial native timber of Brazil. The timber industry is the main economic sector of industrial manufacturing in the Amazon (the first industrial sector still is mining), representing some 15 per cent of the GDP of Pará, Mato Grosso and Rondônia states. In 1998, the gross income of the sector was estimated at US$ 2.5 billion. The timber industry generates approximately 500,000 direct and indirect jobs in the region.

These are significant indicators: compared to the past three to four decades, industrial logging in the Amazon was a relatively small, unimportant activity. As Scholz, I., op.cit., remarks, in the 1960s timber had no economic expression in the economy of Amazonia. The value of timber production was much smaller than the value of production of extractivism (natural rubber, Brazil nuts, etc.). Extractivism and family agriculture, complemented by hunting and fishing, were at that time the main activities that constituted the bulk of the region’s economy.

From the mid-1960s the Amazonian economy began to change profoundly as a result of governmental policies directed towards the economic integration of the region with the rest of the country. It is in this context of policies, of the high rates of economic growth of the region, of the accelerated urbanization of the population, and of the expansion of the agriculture frontier (see items 3.2, and above) that the development of the timber industry and timber exploitation in Amazonia occurs1.

According to IBGE and FAO estimates, the volume of timber (industrial logs) produced in the region has risen from a mere 1 million m3 in the 1950s and 1960s to 2.3 million in the 1970s, and to 11.4 million m3 in 1980. From then on the annual volume of production, as per IBGE and FAO statistics, shows a considerable and steady growth. Other sources of data (regional research and surveys carried out, particularly those by IMAZON op.cit.) also reveal this continuous increasing trend, and place the present (1997-98) annual volume of production in the order of 28-30 million m3 of industrial logs.

In the last decades, the growth in production of native industrial timber in the Amazon not only (i) responded to the increase in demand for this type of raw material due to the rapid expansion of the region’s economy, but also (ii) responded to a greater extent to the demand for native industrial timber from the rest of the country (the “centre-south”) – structural timber, sawnwood, veneers, plywood – a demand that expanded considerably. Note that, in the past (in the 1960s and 1970s), this demand from the centre-south used to have alternative important sources of supply that were earlier constituted by native timber proceeding from the Brazilian Atlantic forest (a source of timber that, from the 1980s, practically ceased to exist due to the shrinking of these forests).

Another source of growth in the timber industry in the Amazon has been international trade (resulting from the increase in foreign demand for tropical hardwoods, combined with a relative decline in traditional sources of supply from countries like Indonesia, Malaysia and others – a decline that had been anticipated in various forecast studies of the international market for tropical timber). Although Brazil, and the Brazilian Amazon, currently still supplies only 4 to 5 per cent of the global market for tropical hardwood (20 years ago this share was 2 to 3 per cent at most), exports increased significantly since the mid-1980s, having grown from US$ 100 million in 1985 to about US$ 400 million in 1997.

As put forward by IMAZON11 “out of those 90 per cent of native tropical timber produced in the country, that is exploited and primarily processed in the Amazon, 85 to 86 per cent go to domestic markets, and the rest, 4 to 5 per cent, go abroad. In the south and southeast of Brazil there is concentrated the biggest and most intensive trade in tropical timber in the world, more than double that imported by the whole of the European Union”. Indeed, this information that Brazil has an enormous consumer market of tropical timber should not be lost sight of.

It is important to underscore this continuing and strong demand stimulus on the expansion of the timber industry in the Amazon. First, because it means that it does and will continue to exert a very significant influence on the expansion of the region’s industry, and, thereby, on deforestation. Second, it must be recalled that in some quite important international documents on forest policy this influence of demand has been played down or underestimated, with implications for Brazil and the Brazilian Amazon. The most outstanding one is the 1991 World Bank Policy Paper – The Forest Sector32 (still in force, it must be recorded) in which, as Lele, U et al.27 remark, “small cultivators and rural producers had been identified as the most important source of deforestation in the Amazon, based on the assumption that much of the urban industrial demand for wood products would be met by imports from temperate countries”.

As the cited authors state, “unlike the Bank’s 1978 Forest Policy, this assumption underestimated the power of domestic urban and manufacturing demand for wood products and its implications for forest policy. Most of Brazil’s domestic needs for wood products have been met from its own forests”. This is really astonishing! Empirically speaking, that assumption is, in the entire economic history of Brazil, just the opposite of what has always happened. Brazil has never been an “importer of wood”. On the contrary, the country has always met its demand with domestic production of wood, and it has always been an “exporter of wood”. Indeed, the exhaustion of the Atlantic forest has a lot to do with that!

Having said that, and for the purpose of providing further figures on the situation or profile of timber exploitation and timber industrial activities in Amazonia, Table 3.16, extracted from the works of Veríssimo, A. et al.28, and Nepstad, et al.19, gives an idea of the distribution of the industrial timber units (sawmills) and their corresponding timber exploitation volumes per state of Legal Amazonia. As will be seen from Table 3.16, in 1997 there were around 2,500 units responsible for demanding or processing a volume of industrial roundwood (logs) of around 28 million m3. Regional processing capacity is very much concentrated in Pará, Mato Grosso and Rondônia.

These 2,500 sawmills are distributed in approximately 75 “poles or regions of timber exploitation” or “logging centres”. According to these authors’ research, these timber poles are responsible for more than 90 per cent of the total production of timber in the Amazon. Each of these logging centres produces at least 100,000 m3 of roundwood (logs) per year. These poles have been circumscribed taking into account the average economic extraction distances of logging that predominate in Amazon areas/regions, on the basis of the interviews carried out by the authors on 1,393 sawmills located in these centres. Mahogany mills have been excluded from the study because their immediate effect on the forests is small compared to other logging, and the volume of mahogany extraction is much less than 5 per cent of total Amazonian production.

The harvest rates (m3 of roundwood timber per hectare of forest) were obtained from the mills’ harvest records, and from there the authors estimated the forest area required to supply each centre’s production. The estimated average rates of logging lie within the interval of 19 m3/ha (low-intensity), 28 m3/ha (medium intensity) and 40 m3/ha (high intensity). Maximum total forest area required to supply the centres was estimated at 1,509,000 ha per year. This would be the area necessary for the production of 28 million m3 of logs in 1997. The authors say that this area of “undisturbed forest” has been logged in 1996/97 and leave the impression that the total “affected”/“logged” area of forests in the Amazon would have been that area plus the year’s total deforested area.

We have seen above (item that INPE’s researchers have rejected this idea: as Krug, T. et al..20 have shown, beyond the area effectively deforested in a year, only 10 per cent at most of this deforested area, on average from 1989 to 1998, would have been affected by selective timber logging. Therefore, in terms of the industry’s demand for timber, the area needed for that purpose would, at most, be made up of the deforested area plus 10 per cent related to selective cutting (the latter being an area that partly regenerates, and partly may be subsequently clear-cut, but not immediately).

Table 3.16: Production of industrial roundwood, per state, logging centres, logging intensity and “forest area affected by logging” – 1996/1997


No. of


No. of



of logs in

million m3

Logging intensity

(% of production)


Necessary area to support production (km2/year)





100 0 0






100 0 0






100 0 0






0 0 100






100 0 0






11 61 28






25 75 0






100 0 0






100 0 0






49 41 10


Source: Nepstad, D. et al.; L=low; M=Medium; H=High

The more than 2,500 sawmills in operation in Legal Amazonia, which extract about 28 million m3 of logs, place the region as one of the greatest tropical timber producers in the world (alongside Malaysia and Indonesia). Most of this timber is transformed into sawnwood and the rest in veneers, plywood and other industrialized timber.

An approximate idea of the size of the participation of Amazonia’s industrial timber (sawnwood, veneer and plywood) relative to the national total (including native forests and planted forests) can also be gathered from Table 3.17, in terms of the consumption of timber by the different forest-based sectors of the Brazilian economy. Assuming that 90 per cent of all sawnwood, plywood and veneer produced from native forests in the country originate from Amazonian forests, this would mean that the region’s share would amount to 60 per cent of the overall national production of these goods (obtained both from native and planted forests).

Table 3.17: Consumption of industrial roundwood (logs) in Brazil, per sector – 2000 (000 m3)


Native forests

Planted forests


% of native forests

Pulp and paper










Indust. fuelwood










Veneer and plywood





Reconst. panels*










Source: STCP;
*Reconstituted panels include particle-board, fibre-board, and MDF.

3.4.2 Deforestation and industrial timber exploitation

Strictly speaking, logging, as practised in Amazonia, cannot be considered as a forest activity on its own. It is, in fact, either a land-clearing operation of forest conversion (clear-cutting) to give way to other alternative competing land use activities (such as crops, pastures) or a partial harvesting of timber (exploitation/felling of the best individuals of valuable commercial timber stocks) that will ultimately also lead, at a later stage, to agriculture or cattle-breeding activities replacing forests. Even “low-impact” highly selective logging, as still practised in some flooded forest areas (varzea forests) in Amazonia, is a predatory and empirical type of exploitation, that may lead to impoverishment of the forest, and even threaten the survival/reproduction of species in some limited areas. With population pressure and the increasing occupation of some remote areas in the Amazon, traditional selective logging tends to be more intensive, and thus becomes as damaging as logging directly associated with deforestation.

“Deforestation” logging in Amazonia ensures conversion of forests to other alternative uses. Actually, logging is but a mere complementary activity, an accessory, to agriculture and cattle-breeding, an operational stage in the process of converting land to subsequent establishment of pastures and crops. As such, logging activity can be a means of financing the conversion of forests to other land uses, and indeed there is ample evidence that this has become so in many areas in Amazonia. Since the end of the 1980s, when timber industry production in Amazonia started to grow considerably, and fiscal incentives to agriculture and cattle-raising and other government-sponsored investments in the region began to decline, this capacity of logging to finance deforestation, as put forward by various of the authors quoted at the beginning of this section, has become more intense in areas of bigger influence of the timber industry, such as for instance in the municipalities of Paragominas-Pará, Sinop-Mato Grosso, and Ji-Paraná/Ariquemes-Rondônia, examples of regions known as timber-industry poles. As pointed out by Reis, E. and Margulis, S.33, until then, the timber industry in the Amazon had a weak influence on deforestation.

As part of the process of deforestation under way in the Amazon region, logging interests are very much the same as those of farmers, shifting-cultivators, rural producers and ranchers. In this sense, loggers’ interests reinforce deforestation interests, and vice-versa. They seem to go hand in hand, bound together by an identity of association as if they were al partners in a business: deforestation. As very well put forward by Carvalho, J.C.34 in his description of the political configuration of the problems of the Amazon rainforest: “When conservation activities and sustainable use of forest resources are ranked as secondary or subsidiary factors in the development process, the vision and focus of the community regarding forest issues are subordinated to its higher ranked day-to-day activities. These are of more immediate interest to the social groups that participate in the productive process and dominate political decisions. In many countries and regions, particularly where ample land is still available, rural development is based largely on the expansion of the farming and ranching frontier, converting forests into farmland, in parallel to the empirical and predatory exploitation of forest resources of greater economic value. In these cases, rural communities do not view forestlands as sheltering resources that can be managed, but rather as strategic reserves of land available for take-over by the farming and ranching process, making merely residual use of timber left over from cleared areas.”

As Schneider et al..6 underscore in their descriptive analyses, the major part of timber industry logging in the Amazon is done as a complement to agriculture and cattle-raising. As a result, the exploitation or logging frontier of the industry has followed the expansion of the agriculture frontier in the whole of the region. The overwhelming pattern of the mode of exploitation is predatory logging for industrial or commercial timber, characterized by excessive damage to the forests, excessive pressure over high valued timber species and greater risk of fires in exploited areas.

This pattern of predatory or “deforestation” logging closely associated to the expansion of deforestation has been exhausting forest resources in all old “timber-pole” regions of the Amazon. By the end of the 1990s, timber industry logging in old frontier areas, such as Paragominas (Pará), Sinop (Mato Grosso), Vilhena-Ji Parana-Ariquemes (Rondônia) was already in decadence and, according to estimates, the scarcity of wood in these regions will be forcing timber companies to migrate or to shut down in the next five years. In frontier areas not so old (intermediate age), as in the north of Mato Grosso state and in Pará (Tailandia-Maraba), the authors estimate that the natural forest timber stocks will be sufficient for an additional 10-20 year period of exploitation under present conditions. In the regions of more recent agriculture frontier expansion (the new frontiers) as, for example, Novo Progresso (Pará), Novo Aripuana-Apui (Amazonas) and Senador Jose Porfirio-Portel (Pará), the estimates are that stocks could last up to 30-40 years.

As per the analyses of Schneider et al., op. cit., and as illustrated by the map on the mobility of the timber industry in the Amazon, elaborated by the authors (see Map 3.15), industrial timber companies and loggers of Rondônia are moving into Bolivia and to Amazonas state, while the timber industry of the old frontiers of Pará and Mato Grosso are migrating to the newer frontiers of western Pará and southeastern Amazonas. As the cited authors state, the current logging model has a strong impact on the economies of communities in the Amazon. Following the expansion period, the consequent exhaustion of resources results in an inevitable economic recession in the local economy. The gravity of this recession depends on the local agricultural potential, that is, on the extent to which the emerging agricultural economy can replace the loss of the timber economy. For example, Paragominas, the oldest logging frontier in Amazonia, established at the end of the 1970s, is confronting a grave shortage of primary material due to the exhaustion of its forests. In the past five years, approximately 50 sawmills closed or migrated, and the volume of wood processed fell approximately 30 per cent. A similar phenomenon can be witnessed in the county of Sinop (Mato Grosso), one of the largest logging centres in the 1980s, and in Redençao (south of Pará). However, due to the fact that Sinop and Redençao are located in areas of open forest (characterized by low density of marketable timber) situated in the dry zone, the decline in timber harvesting has been more rapid than in Paragominas. In Sinop, the number of sawmills fell from approximately 400 at the end of the 1980s to fewer than 100 at the end of the 1990s. However, in this county, the decline in the timber sector has been largely compensated by the rapid growth of agriculture, principally ranching and soybean cultivation”........ “If market forces are not restrained, communities constructed in the latter (newer frontiers) areas during the logging “boom” will become increasingly depressed during the subsequent “bust” to an ever weaker agricultural base.”

Map 3.15: Timber industry poles and migration of industrial logging in Legal Amazonia

Source: Schneider et al. (2000)

It is this “boom & bust” model of timber resource exploitation, associated with deforestation, that conditions the nature and workings of the timber industry in the region, as well as its future prospects. This is confirmed by the work of Scholz, I. op.cit, and on a recent survey of STCP35 on the problems and constraints of the Amazonian timber industry, whose key limitations on the development of this industry may be summarized as follows:

• sawmills operate with used machinery that in general is not renewed. Profits are not reinvested in the mills, but rather in the purchase of lands, or are accumulated to prepare the “opting out”, i.e., to leave the sector in order to invest in cattle-ranching, agriculture and urban activities, mainly in services;

• low productivity, increasing costs of raw material and scarcity of noble species lead to a constant decline in the returns and profits of mills;

• although recovery rates of raw material may have shown some improvement over time, they still fall short of the potential offered by the quality of the raw material (this suggests that the shortage of raw material may not be strong enough, and the cost high enough, to enable the industry to optimize its recovery rates, and/or the industry may not have the skills to do this);

• most timber companies are supplied exclusively from third-party forests. Few invest in the formation of a forest-base of their own;

• since most mills produce for the domestic market only, and this market does not require specific quality norms, the mills expand and survive on this indifference of the market towards quality;

• the introduction of new species has not changed the pattern of behaviour at the mill level: the increase in the number of commercial species has allowed more intensive cutting in areas already exploited (cutting of so-called secondary species, that were not exploited in a first cutting cycle);

• mills are not willing or interested in facing costs, risks and necessary efforts to introduce new species in foreign markets;

• the longevity of milling activity is based on the perpetuity of the predatory model associated with deforestation, changing/moving the site of the mill in accordance with the abundance of forest resources;

• the vision of the timber industry, by the entrepreneurs themselves, as a transitory and itinerant activity becomes a barrier to its development: there is no long-term growth strategy for the companies, of either opening new markets or production diversification.

The statistics for sustainably managed timber production are still deficient and controversial. Item 3.4.3, below, provides a discussion of PMFS under control of IBAMA and of SISPROF. Since the whole of SISPROF is not yet fully operational (particularly the “data bank” module of automated registration of all exploitation authorizations, and the corresponding issuance of the conceded licences for transportation of the authorized exploited timber (“ATPFs” or “Stamps of Origin” – see Sections 4 and 5 below) there still remains a difficulty in establishing more accurately the annual volumes of timber effectively produced from PMFS in Amazonia. For 2001, the potential volume of production from “apt” PMFS could have reached the total of 9 million m3, as shown below. For 1997, the data on annual potential production would be controversial, given the start of an extensive scrutiny carried out by IBAMA of all PMFS in Amazonia in 1996/97.

There are various and different estimates concerning production from forest managed areas. Hummel, A.36 estimates that the timber production under management was around 7.11 per cent of the total production of 28 million m³ in 1996/97. Veríssimo, A.37 reports that SFM covers 5 per cent of the total timber production, and Angelo, H.38 estimates that sustainable timber production was 5.6 per cent of the total timber industry production in 2000. Angelo, H. and Prado, A.C.39 estimated an Amazonian tropical timber supply function for the period 1977-2000, so that the production is a function of the price of timber, managed areas and Amazonian deforestation. The results for the equation are shown below:

Q t = 8,765 + 0,014Pt + 0,114X1t - 0,020X2t R2 = 0,92

(21,11) (4,09) (3,11) (-1,24)

“Q” represents the logarithm of the total tropical timber production in the year t; P is the logarithm of the price index, the aggregate wholesale price index, as a proxy for the timber price; X1 measures the deforestation area in the year t, it is a variable that captures the non-sustainable tropical timber production; X2 is the logarithm of the managed area in the year t.

The adjusted equation explains 92 per cent of the theoretical specification. The variable “managed forest area” does not have the expected sign, and it is not statistically significant at a 10 per cent probability.

Based on these results, the authors concluded that Amazonian tropical timber production is highly dependent on deforestation timber. The expansion of production under forest management must be stimulated by the implementation of mechanisms that differentiate prices of timber (timber from deforestation x timber from managed areas), overcharging deforestation timber and making sustainably forest managed timber more competitive.

3.4.3 Forest management plans in Amazonia

Forest management still is a relatively incipient activity in the Brazilian Amazon. Due to the deforestation process under way in the region and to the continuous expansion of the agricultural frontier, the exploitation of native timber resources is carried out in close association with the frontier’s expansion. The vast majority of the timber industry of the region is supplied by timber from deforestation areas (present and future), i.e. from areas destined for soil conversion to agricultural and cattle-raising activities.

As Scholz, I.31 observes, the Amazonian timber industry was “moulded” to the deforestation process occurring in the region, and to the easy and cheap supply of “deforestation” timber. So, as the quoted author states, and as all research in the 1990s indicates, the timber sector firms are characterized by great instability, having an average life span of seven to 10 years. Such instability is due to the typical cycle of the sawmills, which work until the timber reserves in their neighbourhood have been exhausted, as agriculture and cattle-raising move forward. The existence of timber extraction activity is based on the perpetuation of the predatory exploitation model, moving the sites/locations of the companies according to the abundance of forest reserves.

As per Scholz’s research, the main demands and requests of the timber industry of Pará prove the diagnosis that its main objective is that external control requirements and impositions (mainly from the public sector) on the sector get reduced, so that it can continue in its traditional path of exploitation and timber fellings, together with minimum levels of timber processing, and without having to assume responsibility for the sustainable use of the native forest resource. In fact, the set of incentives that give shape and body to this predominant pattern of predatory exploitation of timber associated with deforestation, and that condition the profile and behaviour of the timber industry in the region, make it extremely difficult for forest management activity to become feasible and diffused in Amazonia.

Even with a greater control of the activity and an improvement in the enforcement of legal regulations and norms (see sections 3.5.1 and 3.5.2 ahead), which require industry to exploit resources through forest management (with the employment of PMFS), which facilitates the verification of exploitation conditions, maintenance and renewability of the resources, effective harvesting of timber from these plans and/or areas is not only marginal in terms of total regional production, but also some of them may be fictitious or irrelevant. As a result of this, several PMFS may be used as a “façade” for legalizing timber that has been irregularly obtained from other areas (see section 3.5.2, in particular).

Several measures have been adopted on behalf of PMFS in the Amazon and are described in Section 5. Although insufficient to compensate or to remove existing disincentives to forest management in the region, some of the measures have adopted the more systematic monitoring and technical auditing assessments of PMFS implemented by IBAMA in Amazonia since 1996/97; these have contributed to “moralizing” the PMFS instrument. In that year, after the first, more intensive and far-reaching technical auditing undertaken by IBAMA on all registered PMFS, out of a total of 1,592 management plans that were audited, only 484 were considered “apt”, 992 were suspended and 116 were legally cancelled (see Tables 5.6, 5.7 and 5.8, in Section 5). Since then, and as described in Section 5, IBAMA has been implementing a set of actions of control and technical support to PMFS in Amazonia. In 2001, as pointed out in IBAMA’s “Report of monitoring and technical auditing on PMFS”40, and as a result of the field and office assessments, out of a total of 1,059 audited timber PMFS (21 plans were non-timber, viz.,“heart of palms”), 549 were considered “apt”, 459 were suspended (of which 48 were soon legally cancelled) and 51 were “in recomposition”.

Table 3.18 shows the situation, in terms of the number of registered PMFS, area of management under annual production (according to POA), and corresponding potential volume of production.

Table 3.18: Timber PMFS – Total number of audited plans, annual harvesting area and potential volume of production – per state of Legal Amazonia – 2001


Number of plans

Harvesting area

(in ha)



(in m3)

Average volume

Per ha





















Mato Grosso

























Source: IBAMA

As will be seen from Table 3.19, taking into account “apt” PMFS, in 2001 there was an annual potential volume for extraction, capable of being authorized, of 9.3 million m3. This represents approximately 30 per cent of the total regional industrial roundwood (logs) timber production which is 30 million m3 (the extraction volume is the requested volume in the POAs, and may end up either not being fully authorized or not being fully practised by the producers).

In terms of potential annual area and corresponding production under forest management in the region, according to audited and “apt” PMFS and their POAs, there would be approximately 340,000 ha corresponding to 549 forest management units (forest estates). Considering that 411 PMFS were suspended, and that out of these a good part could solve the pending problems identified by the audit visits, and that 51 were in “recomposition”, there could be a total of some 1,000 plans operating in a regular situation in Amazonia in 2001. Using this basis for comparison, therefore, one may say that the maximum area of forests under regular forest management production in Amazonia would hardly have reached those 600,000 ha.

As for the quality of forest management at unit level, IBAMA is already capable of reporting on this too, thanks to SISPROF. The next chapters are dedicated to describing and explaining the whole of SISPROF. Here a brief account is given of how PMFS performed, based on the 2001 technical auditing carried out by IBAMA.

Table 3.19: “Apt” timber PMFS – number of plans, annual harvesting area and potential volume of production – per state of Legal Amazonia – 2001


No. of plans


area (ha)

Production volume

(000 m3)




Other plans


Other plans






























M. Grosso



































Source: IBAMA

Details of the system of “indicators and verifiers” of PMFS, defined in SISPROF, may be found in Section 5 of this document. Just as an example, Table 3.20 shows the “quality” results of “apt” timber PMFS in highlands of the Amazon (this category corresponds to the majority of PMFS in Amazonia – only in the state of Amazonas is there a bigger proportion of lowland (varzea forests) timber PMFS, and management in varzea forests is relatively inexpressive in the other states of Legal Amazonia).

As Table 3.20 shows, only the states of Pará and Amapá, with more significance, have shown general averages above a level considered as regular. Mato Grosso has shown the smallest average quality index among all apt PMFS in the region. In relation to activities considered in isolation, those related to the “general aspects” have had the least attention, although the averages of the other groups of activities have not shown much difference between them. The group that had the best average was that of pre-exploitation activities. This could, perhaps, be expected since these activities are more easily transmitted in forest management than the further activities (exploitation and post-exploitation) that need better infrastructure in order to be more adequately passed on and absorbed. It must be remarked that these assessments, now made possible by SISPROF, are in their initial stages of use in IBAMA, and therefore will need further refinement. But, in particular, the verification system will need more time to be systematically applied, in order to allow better and more relevant comparisons.

Table 3.20: Quality of “apt” timber PMFS in Amazonia (high land, individual, industrial scale and regime of area control) - 2001






Média Geral


Labour security

Camping infrastructure

Monitoring of activities

Delimitation of AMF/UPA

Orientation tracks

FI 100%






Patio operations

Silvicultural treatment

Forest protection

Infrastructure maintenance

Monitoring of forest development






































































































































































In terms of the general pattern of different land uses in the region, as we have attempted to describe in the previous sections, the above figures on forest management as a permanent land use activity carried out in Amazonia clearly reveal that the activity is no more than a “grain of sand” in the whole regional landscape: in a universe of at least 390,000 rural properties with a corresponding area of 180 million ha (according to INCRA’s 1998 data), there would be only around 1,000 properties (0.33 per cent) at most, dedicating themselves to forest management. Or, if we take the figures from the IBGE 1996 agricultural census, out of a total area of 116 million ha of more than 500,000 “agricultural establishments” reported in Amazonia, whose area under some form of use (agricultural, pastoral and other uses) would be nearly 62 million ha (see Tables 3.5 and 3.7, under item, and where the latter table shows that 77 per cent, 6.8 per cent, 1.5 per cent and 9.5 per cent account for cattle-raising, temporary crops, permanent crops and “abandoned areas”, respectively), the participation of the area under forest management production would be, at most, 1 per cent of that total area under use by all agricultural establishments in the region.

Note that Table 3.7 also shows that “other types of land use in agricultural establishments, which refer to lands in rest, planted forests, extractivism, etc.” amount to 4.6 per cent of the total area in some form of agricultural use. Therefore, since the agricultural census does not have specific data on forest management area under production, or in PMFS, what one can do is to guess that the real participation of areas under forest management must be included in those 4.6 per cent, and that the participation would, at most, amount to the 1 per cent order of magnitude that we have here estimated for comparative results only.

Another interesting indication of the present smallness of “private permanent productive forestry” in Amazonia may be gathered from data provided by WWF/Brazil41, which states that some private companies committed to the sustainable use of their forests, and with an increased valuation of their products, have been seeking to obtain certification for their managed forest areas. According to WWF/Brazil the total certified area in the Amazon is currently 353,313 ha. Just for information, and as anyone would easily guess, PMFS are also concentrated in the region of the “deforestation arch” of Amazonia.

Thus, in terms of the above numbers, one can see quite well why a private permanent productive forest management activity in the Brazilian Amazon does not yet have any reasonable significance in all the interests and aspirations of the rural population of the region, be it from communities (in the total of audited PMFS in 2001, only 11 belong to the communal category), be it from small settlers or family croppers, or be it from bigger rural producers, agriculturalists or cattle-ranchers. This profile on the tiny size of the activity shows the gravity of the political challenge that will have to be faced in trying to convince the rural landholders of the Amazon that their properties (i.e. particularly a substantial part of them, viz. their legal forest reserves) must be maintained and destined for an economic activity which, until now, is practically unknown or non-existent in their properties, namely, SFM.

Although some farmers and communities may be starting to manage their forests, instead of contracting tree-felling to outsiders, or selling timber-cutting rights (both linked to a present, or future, subsequent conversion of their areas to agriculture), in fact very few farmers in the Amazon perceive the opportunity of their legal forest reserves (much less of their possible increase). Very few perceive this opportunity and, moreover, very few possess the necessary capacities. As a consequence, the legal forest reserves are poorly observed and owners see them as a constraint rather than an opportunity. If SFM is not widely diffused, decisively supported and motivated, in order to demonstrate that it could be an economically attractive and competitive land use activity, it will hardly succeed in Amazonia. Moreover, this also shows how unrealistic it may be to propose and expect that all of these properties maintain and keep their legal forest reserves untouched, for purposes such as that of “an integral protection of bio-diversity” (as environmentalists usually propose, even if the proposal is genuinely accompanied by some, though still really non-existent, broad mechanism of financial or economic compensation).

On the other hand, what all rural landholders (small, medium and big) of Legal Amazonia know quite well is the activity of deforestation and its corresponding income generated from timber sales, that allow them to finance the conversion of their forests to other agricultural or cattle--raising uses. It is this activity, indeed, whose immediate product/output (deforestation timber) is not adequately captured in the production statistics of the agricultural census (even though, in fact, it does not really mean “use” or “output of a resource use” but “disuse”/elimination of a resource, something that is naturally and generally not reported by informants to the agricultural census enquiries) that is familiar to and interests all rural owners.

It is expected that with the full and permanent functioning of SISPROF, forest management activity will become more transparent and PMFS will be less vulnerable to being used as a “fictitious” instrument. One could expect as well that the hard-data on deforestation – viz. the exploitation and recovery of timber from deforested sites – per unit of landholdings in the Amazon, together with other information on the landholdings, such as the corresponding registry and maintenance of legal forest reserves, will become more readily available and more effectively controlled.

3.5 “Command and control” of deforestation and of forest resource use and access: progress of regulation and enforcement

3.5.1 The evolution of legislation

The Brazilian forestry sector is extensively regulated along all its production and commercialization chain. Basically, the regulations that guide deforestation and the exploitation of natural forests fall within the following areas: (i) environmental impact assessments; (ii) deforestation and burning authorizations; (iii) cutting restrictions specific to rural properties; (iv) cutting restrictions specific to species, regions and ecosystems; (v) technical obligations and requirements of SFM; (vi) conditions/obligations of forest replacement and recomposition; (vii) conduct restrictions and requirements of licences for transportation and commercialization of forest products; (viii) restrictions on maintaining legal forest reserves and permanent preservation areas in private rural properties; (ix) rules concerning public conservation units of sustainable use. The basic laws that are the basis of these regulations are the:

Forest Code (Law 4771 of 1965 and subsequent alterations)

National Environment Policy (Law 6938 of 1981 and subsequent alterations)

Law of Environmental Crimes (Law 9605 of 1998)

Law of the National System of Conservation Units (Law 9985 of 2000)

The principles of environmental protection have been consolidated in chapter V, article 225 of the Federal Constitution (1988). The above federal laws are complemented by specific state-level legislation. A series of lower-rank legislation such as federal decrees, national environment council resolutions, normative instructions of the Ministry of the Environment and resolutions of IBAMA converts the principles and commands established in the federal laws into regulations and rules to be enforced in practice. At state level a similar process is followed.

The main dispositions and rules of the above-mentioned legislation that are more directly concerned with the regulation of deforestation and of access to forest resources in Brazil, and in the Amazon, are highlighted with the purpose of providing grounds for understanding how this legislation conditions the design and workings of enforcement mechanisms such as monitoring and control systems like SISPROF.

National Environment Policy Law

The National Environment Policy Law, and CONAMA Resolutions 001/86, 011/87, 006/88, 009/90, 010/90 and 237/97, establish the principles, rules and obligations for the environmental licensing of activities that are potentially pollutant or degrading to the environment. Agriculture, animal husbandry and agrarian reform settlement projects are land use activities listed in Resolution 237/97, leaving to the states of the federation the competence to adequate licensing norms to their peculiarities. In relation to these land use activities, the objectives of licensing are, among others, to protect and monitor permanent preservation areas in rural properties that request deforestation permits, reduce the processes of soil erosion, the pollution of waters with agrotoxic substances and avoid forest fires. States like Mato Grosso have built modern monitoring systems of rural licensing on the basis of specific legislation enacted for this purpose (see Box 3).

Forest Code

The 1965 Forest Code has been revised a few times. In 1989, just after promulgation of the 1988 Constitution, under the aegis of the first more comprehensive programme of the Federal Government for the environment – “Our Nature Programme” – some important changes were introduced in the Forest Code by Law 7803/89. Among them the following may be noted:

(i) rural properties have been compelled to register their obligatory legal forest reserves in land registries at notary publics;

(ii) clear-cutting was explicitly prohibited in legal forest reserves of rural properties;

(iii) legal forest reserves could not have their destination altered in cases of “inter-vivos” or “causa mortis” transmissions/transfers, or in cases of dismemberment of areas.

(iv) the obligation to recompose legal forest reserves in rural properties was imposed by the new agricultural law passed in 1989;

(v) although article 15 of the Forest Code had since 1965 prohibited the exploitation of primitive forests of the Amazon basin under empirical form, and that these forests could only be used by observing technical plans of conduct and management to be established by act of the public power within a year - and until 1989 such regulation had not been issued - a new article was included in the Forest Code by Law 7803/89 establishing that “the exploitation of forests (in any region of the country), public or private, would depend on the previous approval of IBAMA, as well as on the adoption of techniques of conduct, exploitation, forest replacement and management compatible with the varied ecosystems formed by the forest cover”;

(vi) some limits on the strips of permanent preservation areas, either public or private, were modified, such as banks/margins along water courses and around water sources.

The other main dispositions of the Forest Code concerning deforestation and use of forest resources and products were maintained. These were:

(i) permanent preservation forests in rural properties must be observed on riverbanks, steep slopes, hilltops and around lakes, ponds and water sources. The Forest Code has put indigenous reserves at the same level as permanent preservation forests;

(ii) forests (and other wooded vegetation) that were not subject to any limited regime of utilization, and except those of permanent preservation areas, could be clear cut, in rural properties, provided the owner/holder of the land maintained at least 20 per cent of the area of the property with forest cover as legal forest reserve (in all regions of the country except in the Amazon) or at least 50 per cent of the area of the property with forest cover as legal forest reserve, in the case of the Amazon region;

(iii) obligatory forest replacement for industrial consumption of native forest raw material (timber, charcoal, fuelwood and other forest raw materials) that exceeds the consumption from rationally managed and exploited forests of the consumer and/or from forests owned by others but with whom the owner had supply contracts;

(iv) among the penal contraventions established in the Forest Code the following could be noted: the destruction or damage to permanent preservation forests; the cutting of trees in permanent preservation forests without permission of the competent authority; to set fire to forests; to transport or keep timber, fuelwood, charcoal and other forest products without a valid licence, issued by the competent authority, for the trip or storage; to receive timber, fuelwood, charcoal and other forest products without requiring a valid licence of the seller, and without keeping the copy of the licence that must follow the product until final manufacturing processing.

Note that unauthorized or illegal deforestation or clear-cutting of forests, other than permanent preservation forests, was not considered as penal contraventions.

(v) extraction of fuelwood and other forest products, or the production of charcoal, are free in planted forests not considered as permanent preservation forests;

In 1996, together with the disclosure of INPE’s 1994/95 data on deforestation of the Amazon, the government decided to send to national congress a specific “provisional measure” (which until today is under discussion in congress) that alters some dispositions of the Forest Code (the modifications thereby introduced have been in force since then, despite not yet being ratified by Congress), and provoked the opening of a wide and substantial debate among Brazilian society about the need to proceed to more significant changes in the norms and rules that regulate the use and access to lands and forest resources in the Amazon (and in the rest of the country). The following changes, worthy of mention here, have been promoted by the measure:

(i) a strong legal limitation on the alternative use of land in rural properties that are covered by forests in Legal Amazonia. Legal forest reserves in rural properties in Amazonia increase from 50 to 80 per cent minimum of the total area of the property in the case of forest cover, and from 20 to 35 per cent minimum in the case of cerrado vegetation cover (no change is introduced in the percentages for the other regions of the country, thus remaining at 20 per cent);

(ii) the prohibition of conceding deforestation permits/licences if the property already has a deforested land which is underused, degraded or abandoned;

(iii) the possibility of compensating deficits of minimum legal forest reserve in a property in exchange for another area in a reserve-surplus property (but only in the same water-basin and in the same state);

(iv) the possibility that these compensations be made with “bonds/equities representing legal reserves”, and not only through unilateral or bilateral transactions of physical areas;

(v) the possibility of constituting “non-obligatory forest reserves” such as “forest serfdoms” and others, that will be able to be negotiated in a “market for compensations” for forest protection.

Since 1996 this provisional measure has been extended by government 67 times, and the bill is still awaiting final decision in Congress, but no agreement has so far been reached. The measure has provoked tension and reaction from rural agricultural producers (small family producers, small agrarian reform settled croppers, medium-size farmers and big rural producers) in the whole of the Amazon who, because of an absence of other mechanisms or incentives that reinforce the conservation and sustainable use of forests, still do not perceive that they may gain with the new proposals (small croppers and rural producers in the Amazon, whose properties are smaller than 150 ha have managed, through lobbying in Congress, to introduce an exception to the 80 per cent legal reserve rule: they succeeded in getting a modification that maintained, for their property size, the former percentage of 50 per cent). Indeed, very few farmers and communities perceive this opportunity, and even possess the necessary capacities to make best use of their legal forest reserves. As a consequence, these legal reserves are poorly observed, and owners see them as a constraint rather than an opportunity. As has been mentioned above, only a few farmers and communities are managing their forests instead of contracting harvesting to outsiders or selling harvesting rights in forest-land conversions.

Law of Environmental Crimes

In February 1998 an important step forward was achieved in terms of enforcement of environmental policy in general: the passing of Law 9605 (the Law of Environmental Crimes), which established and clearly defined the crimes (and respective penalties and fines) against the environment. The Law established the crimes against fauna and flora, those related to pollution and other environmental crimes, to urban space and cultural patrimony and crimes against environmental administration.

The dispositions of Law 9605 relating to the crimes against flora (Section II of the Law) are underscored here. As disposed in the Law, the following actions are now crimes punishable by specific penalties and fines:

(i) destroying or damaging permanent preservation forests;

(ii) cutting trees in permanent preservation forests without the authorization of the competent authority;

(iii) causing damage to conservation units;

(iv) provoking fires in forests or woods;

(v) extracting minerals, sand, stones from forests of public dominion or from permanent preservation forests;

(vi) cutting or transforming commercially valuable timber into charcoal for industrial, energetic purposes or for any other economic exploitation, in disagreement with legal determinations;

(vii) receiving or purchasing, for commercial or industrial purposes, timber, fuelwood, charcoal and other products of forest origin, without requiring the licence of the seller issued by the competent authority, and without the document/copy that must accompany the product until final processing;

(viii) whoever sells, exposes for sale, keeps in deposits/stores, transports or keeps timber, fuelwood, charcoal and other products of forest origin, without a valid licence for the duration of the trip or storage will incur the same penalties as those in (vii).

Note that what once was penal contravention of the Forest Code has now become a crime under the Law of Environmental Crimes. Criminal offences are punishable by detention or imprisonment and/or fines.

Note also that unauthorized or illegal deforestation or clear-cutting of forests, other than permanent preservation forests and/or in conservation units, are still not considered as criminal offences. So, a small cropper, or a big rural producer, that clear-cuts an area (not of permanent preservation forest) of his estate, for the establishment of agriculture or cattle-raising, without a licence or authorization, will not legally be a criminal, provided he is not caught selling timber from his deforestation. If he sells the wood without a deforestation licence (which allows him to legalize his timber for sale) and gets caught, then he could be criminally prosecuted for that sale, and not for the deforestation itself. That, of course, does not exempt any such infractor from other penalties such as embargo on his activities, seizure/confiscation of machinery, equipment and merchandise and other administrative prosecution. But, provided he does not commercialize, transform or process that timber into any sort of economic exploitation, he could not be prosecuted as a criminal.

Note, also, that any buyer (or further buyer along the production and commercialization chain) who buys and/or transports timber, fuelwood, charcoal or any other forest product from an illegal/unauthorized deforestation site will always be subject to prosecution as a criminal.

These features/dispositions of the Law of Environmental Crimes, and also of the Forest Code, seem to put a heavier demand on the need for control instruments to have a series of documents/proofs that unfold or accompany the various steps/links of the production and commercialization chain of forest-based products. Indeed such various “step-wise” documents are indispensable proof for any legal court case or discussion. These requirements of the two Laws also serve to show why the exploiters that buy timber-harvesting rights, transporters, merchants and manufacturers of timber and forest-based products are those that somehow have to be “more fully documented” in terms of their activities. The fines foreseen in Law 9605 have been stipulated in Decree 3179 of September 1999. The Law allows for much higher fines than in the past.

National System of Conservation Units

The issuing of Law 9985/00 has been a significant achievement of environmental policy in Brazil. Besides having clearly defined and established the roles, categories and grouping of conservation units in the country, thereby improving considerably what had been foreseen in this respect in previous legislation, particularly in the Forest Code, the Law (i) fixed the discipline for the charging of different fees/duties related to access and various uses of these units, and defined how funds and other receipts should be used to consolidate them, (ii) set basic rules for the indemnity or compensation of private lands involved in the transformation of these estates into a public conservation unit, (iii) specified the actions and limits to which encroachment near and around the units could be accepted and controlled, (iv) defined penalties and fines for damages against the units, and (iv) established other compensation forms for environmentally damaging actions, in general, whose debts could be legally paid or offset through contributions to or investments in the unit. The Law also established the basic distinction of two groups of conservation unit, namely, the group of strict protection units (national, state and municipal parks, biological reserves, ecological stations, etc.) and the group of sustainable use units (national, state and municipal forests, extractivist reserves, reserves of sustainable use, etc.).

Other regulations on deforestation and forest-based activities

The dispositions of the Forest Code, as well as the extensive complementary array of their regulations, are generally in direct conflict with the economic incentives that drive the private sector (either rural producers as well as timber industry and deforestation logging activities in the Amazon), which makes enforcing these apparatuses, a difficult task. Regulations, also, are rarely targeted at specific externalities.

In 1994, the Federal Government issued Decree 1282/94 and thereby established for the first time, after 29 years, the first regulations of Article 15 of the Forest Code, stipulating the basic norms and principles for the sustainable management of Amazon forests. Decree 1282 would be further amended by Decree 2788/98. These decrees set the guidelines for formulating and presenting PMFS to IBAMA, as well as the ways in which legal forest reserves could be used through SFM. They also reaffirmed, in the case of conversion of forests for alternative uses of land in the Amazon, that deforestation through clear-cutting could only be done if legal forest reserves were maintained, and that it would only be allowed with a deforestation authorization that had to be issued following a previous technical visit to the site by the competent authority.

Norms for the accomplishment of obligatory forest replacement were likewise re-addressed and made less complicated, but considering the increasing demand for forest products and the lack of incentives for replanting, those prescriptions also could not be easily complied with given the continuous and further moving of the agricultural frontier. Requirements for obligatory forest replacement have been particularly difficult to comply with or enforce.

IBAMA or state environmental agencies (provided federative pacts were firmed) would have to approve forest land conversions involving clear cuttings/deforestation. They would also have to approve PMFS (so far, in the Amazon, only IBAMA has been approving such plans).

Transportation authorizations of forest products (ATPFs) are required and are issued on the basis of approved deforestation requests/projects or on the basis of approved POAs of PMFS. Requirements for the authorization of deforestation have been much easier and simpler to fulfil than the requirements for the approval of PMFS or of their annual POAs (though, as is described in Section 5 below, a lot of improvement, streamlining and simplification concerning requirements for the approval of PMFS has been achieved).

Although the Forest Code, and all of its pertinent regulatory legislation, establishes a series of restrictions on the suppression and use of forest resources in a rural property (that permanent preservation forests cannot be exploited and that legal forest reserves must be maintained in each property are major commandments), the basic right to deforest a part of the estate is ensured to each owner or holder, and therefore the right to exploit/cut the forests and trees of that area, without much further restriction, except that the owner and the buyers of his harvesting rights, as well as transporters, get the required licences/permits.

To be more precise, the norms of IBAMA, or of the Ministry of the Environment, related to deforestation authorizations have generally established, to a lesser extent, a maximum volume of commercial timber per hectare that may be authorized/conceded to the owner/holder of the property without a forest inventory, and in an upper extreme a maximum volume (that may be higher than that established for the lower extreme) that could only be determined by a forest inventory. As a rule, in general, IBAMA has tried to maintain authorized volumes per ha at lower levels for deforestation authorizations, comparative to authorized per ha volumes of PMFS. But there have been exceptions to this pattern .

Insufficiency of existing legislation towards SFM

Despite considerable progress reached in the last decade in enacting new legislation that certainly improved the country’s possibilities of better protecting and conserving its forests, there still is a crucial need for specific and more efficient legislation capable of promoting the sustainable management and use of the country’s natural forests, particularly those of the Amazon.

It should be noted that, in the case of deforestation authorizations and control, so far a principle, norm or rule has not yet been established – that either (i) charged for the “scarcity rent” of the forest resource, or (ii) imposed a quantitative limitation on the share of each agent in the supply of the market – that could constrain an owner’s ability or capacity to effect, via deforestation, the reduction of the total availability or stock of the forest resource in the region, because of the negative externality that he inflicts on the other agents for eliminating part of the resource. Note that this type of externality seems indeed to be the case in Amazonia.

As Seroa, R. and Ferraz, C.42 point out, the way in which land property or possession rights get established in Amazonia characterizes a free-access regime to forest resources of the region. In the process of occupation and agricultural frontier expansion in Amazonia, the traditional form of occupying lands through deforestation, with the consequent suppression of forest resources, as the authors put it, shapes a typical model of free access in which the scarcity of the forest resources is not perceived, and after a time leads to their exhaustion.

Property/possession rights are defined “ex-post”, that is, economic agents move to the frontier, convert forests, sell (or burn) timber, initiate an agricultural or cattle-breeding activity, and then wait for the land-title. This, as we have pointed out previously (see item above) is very much conditioned by the rules of Brazilian agrarian legislation for getting property rights on deforested land.

As Seroa and Ferraz, op. cit., remark, in a free-access regime like this, each agent acts on the presumption that it is not worth foregoing an extraction of the resource today in exchange for a bigger stock in the future, because the resource would be captured by another agent, given free access. This is aggravated in Amazonia, because the way of ensuring the possession of land (reducing the risks of losing it or of being invaded) is to deforest it as soon as possible. In the land occupation process a land-holder knows that his neighbour will adopt such behaviour, which means that the action of one agent forces the exhaustion of the resources to the other, thus typifying the mentioned externality.

In this traditional process of occupation and privatization of lands and of forest exploitation in the Amazon, access to forest resources is also free for others that exploit them (loggers, timber industry) without these agents necessarily being the owners of the resources.

So far there has been no limitation on entry, or on access to forest resources that can originate from deforestation, either in terms of the number of extraction agents or timber companies that would exploit or consume the resources in their production processes, or in terms of their having to pay a “cost of use” for using the resource (except for the obligation to register with IBAMA, get the necessary licences and, in the case of the consuming industry, also the obligation of forest replacement. Note that, in the past, the “forest replacement fee” that was optionally paid by consuming industries of timber originating from deforestation was such that its value only included the unit costs of replanting the wood/tree. The “fee” has never foreseen a value relative to the “stumpage value of the resource”. On the other hand, the obligation to physically replant the suppressed timber has been particularly difficult to enforce and to comply with, which indicates that “full resource cost” recovery through this modality has hardly been achieved either).

What must be underscored from these remarks is that the institutional norms and rules so far enacted/available in Brazilian legislation for controlling deforestation and associated forest exploitation are incapable, or insufficient, of restricting access and promoting the exploitation and use of forest resources in a more efficient way. In other words, insufficient in providing ways/instruments for really economizing on the use of the resources.

In addition, this is particularly perverse for SFM activity in the Amazon: since present norms and rules cannot limit deforestation and associated timber exploitation more efficiently, and thereby the free entry of this wood into the market, this reduces market shares and competitiveness of sustainably managed timber in the total market of timber originating from the natural forests of Amazonia.

3.5.2 The evolution of enforcement

Various important policy measures have been taken by the Brazilian Government since the end of the 1980s – a period that immediately preceded the United Nations Conference on the Environment and Development (UNCED/1992) and consolidated in the so-called “Our Nature Programme” (1988/89), considered as a key change in the position of the country in dealing with environmental issues related to the development of the Brazilian Amazon. Among the main measures that were adopted then, the following can be recalled: the suspension of the fiscal incentives of the investment fund for the Amazon (FINAM/SUDAM) for agriculture and cattle-breeding projects in areas of primary tropical forests; the elimination of subsidies to rural agricultural credit; the extinction of big public investment and infrastructure programmes in Amazonia, such as the POLAMAZONIA, PIN/PROTERRA and others; the end of the unified minimum support prices for agriculture; the creation of IBAMA; the creation of various “extractivist reserves” and national forests in Amazonia; and the modifications introduced in the Forest Code, as have already been highlighted.

In the case of control and enforcement, the creation of the “Operação Amazonia”/Operation Amazonia programme constituted the first and more comprehensive set of actions concerned with monitoring, control and inspection of environmental impact activities in the Amazon, such as deforestation, mining, placer-mining, timber exploitation and others (vigilance of frontiers, etc.). The PRODES programme, implemented by INPE since 1988, for example, was created under the auspices of “Our Nature Programme”. From 1989 to 1994/95 much emphasis has been given to the “command and control” instruments with the purpose of seeking to “hold” the progress of deforestation and the somewhat “unruled” wasteful exploitation of forest resources in Amazonia. From 1996, the government has proceeded with still more impetus in the enforcement of regulation and control measures, thanks to new legislation that was enacted but also thanks to further and specific programmes formulated for that end. Programmes such as the “PPG-7 Pilot Programme for the Protection of Brazilian Tropical Forests” have had more latitude of implementation.

Other programmes like, for example, PREVFOGO (Programme for the prevention and combat of forest fires and burnings) and PROARCO (Programme for the inspection and policing in the region of the deforestation arch in Legal Amazonia), besides a series of “annual operations of inspection and control”, involving IBAMA and OEMAs, destined to reinforce the control and monitoring of deforestation and timber exploitation, have been allocated substantial human and financial resources for structure and capacity building, training and purchase of modern equipment, in order to improve the effectiveness of control and inspection actions and the integration among federal and state institutions.

The system of monitoring and control built by FEMA/MT (see Box 3) is a concrete example of these efforts. FEMA has received resources from PPG-7 and from the PRODEAGRO/MT ecological-economic zoning programme to build its licensing and monitoring capacities. Such programmes in other states of Amazonia have also been reinforced under the aegis of PPG-7.

Since “Our Nature Programme”, and particularly from 1995/96, the areas of monitoring, control and inspection of deforestation, forest exploitation and environmental licensing of agricultural activities have been receiving the biggest sums of federal budget resources and of international funds destined both for the Federal Agency-IBAMA and to OEMAs. More recently (2000), the government sanctioned Law 10165/00 that created the “fee for environmental control and inspection”, a specific levy to be paid by all activities of potential environmental impact and users of natural resources, and whose funds will revert to IBAMA and will reinforce the institution’s capability of exercising its policing and inspection competence. The involvement and participation of NGOs (national and international) have grown considerably and have been fundamental in contributing to all of these government initiatives and efforts towards a more decisive action in controlling activities in the Amazon. In 1996, various reports from NGOs on illegal activities in the Amazon involving timber exploitation were published, nationally and worldwide. At that time, the government, as a result of an intergovernmental working group coordinated by the then Secretariat of Strategic Affairs of the Presidency of the Republic, estimated that approximately 80 per cent of all wood that was traded in Amazonia would have been obtained either illegally and/or without the required licences and documentation. Until the mid-1990s the costs of the “illegality” involved in activities of forest conversion and associated deforestation, as well as of timber thefts from indigenous reserves and some conservation units in Amazonia, were still very low.

It is only since the second half of the last decade that this situation began to change more decisively. The enacting of a new legal apparatus (especially Law 9605/98) and the strengthening and increase of inspection and control operations in Amazonia have resulted in quite relevant infringement findings, the issuing of substantial fines and numerous apprehensions of illegal timber. In the state of Pará, for example, the total value of fines imposed by IBAMA in 1999 reached at least US$1 million. In the following years, with the intensification of inspection and policing operations in the region, the total value of fines increased further, and in 2001 may have reached the sum of US$ 15 million for the whole region. By 2001, the total volume of timber confiscated by IBAMA in Amazonia is said to have been 295,000 m3 (cf. Simula, M.43).

These renewed and strengthened inspection and control actions by environmental agencies in the Amazon seem to have led to a considerable increase in the costs of engaging into and maintaining illegal/fraudulent activities in exploiting and trading timber in the region. Even if fines imposed end up not being fully paid, the infringement evidence and indictment represents an additional cost to the infractors due to the need to seek legal advice, the possibility of having assets frozen, loss of part of the raw material, etc. Even though these more effective controls and inspections cause an increase in management costs, the general tendency for agents is to choose the legal path of action.

In general, people tend to continue committing illegal acts as long as illegality costs remain lower than the costs of getting legal timber. One result of the various control and inspection operations of IBAMA and OEMAs in Amazonia was that the costs of remaining illegal were significantly increased. According to estimates of Friends of the Earth/Brazil (FoE)44, because of these stronger and more systematic control and inspection actions, the costs of illegality increased to a level around US$ 4-6 per m3 of logs, something very close to the unit cost/m3 paid to landowners/settlers for standing timber/harvesting rights.

The modality of timber exploitation through PMFS has been, perhaps, the one to suffer more control and inspections. Prior to the beginning of the renewed actions of control and inspections by IBAMA, it was known that PMFS had been widely used to cover up illegal/fraudulent extraction of timber. The streamlining of the rules for SFM and the systematic operations of auditing visits, control and inspection of the PMFS served to “clean” or “moralize” the instrument to a good extent. In 1996/97 IBAMA started a wide and complete auditing review of PMFS in Amazonia. Out of around 2,800 PMFS that were registered, and whose monitoring/control visits started in that year, 1592 plans were audited in the field (the rest were suspended or cancelled through mere documentation or legal review) and only 484 were considered “apt” to operate.

Box 3

System of inspection, licensing and monitoring of rural properties of Mato Grosso State

Since 1998 the state of Mato Grosso has instituted an LAU (single environmental licence) for rural properties to facilitate the licensing of agricultural activities, authorizing at once the location, establishment and operation of land use activities.

The state, through a Federative Pact signed with IBAMA, has the right to authorize deforestation in rural properties bigger than 1,000 ha (which amount to 85 per cent of the area of rural land holdings in the state, but which represent a small proportion of their number). The deforestation authorizations that are processed/operated by FEMA are issued on the back of the image-chart of the rural property that applies for the LAU. The system adopted by FEMA for the control of deforestation and of technical visits to the sites uses information technology systems, remote sensoring, geographical information systems and global positioning systems. FEMA promotes the production of information as much as it can. The map of annual deforestation for the state is done by private companies of the geo-processing sector. FEMA also uses private professionals of the environmental licensing area. FEMA only monitors, audits and inspects alterations in the use of land at the level of rural properties. The monitoring, control and inspection of timber exploitation is the responsibility of IBAMA, that is, the authorization/concession of timber exploitation volumes and consequent ATPFs are issued and conceded by IBAMA. deforestation authorizations of smaller areas in the state also continue to be IBAMA’s responsibility.

On the basis of updated image charts (1:50,000) FEMA identifies in the deforested areas bigger than 1,000 ha the deforestation that occurred in APPs (permanent preservation areas). On these images, the inspection teams move to the field for the inspections, using also image charts of the previous year, to check whether or not landholders have deforestation authorizations. Each team is formed by one graduate technician and a driver. At least two teams have been mobilized to audit each property (the presence of two vehicles is important given the risk of break-downs and “atolamentos” – that is, getting stuck in the mud).

The work of FEMA has had the financial support of PPG-7 .

Rural landholders have to provide a digital map of their properties, including the limits of the legal forest reserves of these properties.

By 1998 the technical team of the Directorate of Forests – DIREF of IBAMA - had also begun to design and take the first steps in establishing SISPROF. Another proposal, though much less comprehensive, for an integrated control and monitoring system for forest exploitation and transportation of forest products in Amazonia would later (2001) be made by IMAZON (see Box 4).

The results of the auditing actions in PMFS, together with the other control and inspection operations launched by IBAMA in Amazonia since 1997, must have contributed to a reduction in the relative freedom to which timber industry operation and timber exploitation associated with deforestation in the Amazon were accustomed.

By 1999, the context of the considerable progress achieved by the government in agrarian reform policy, in terms of the numbers of settled families in the region (a boastful result of the government’s achievement in social rural policy) started to impose pressure on other spheres of government, particularly pressures for relaxing or softening of some of the rules for controlling and authorizing deforestation.

The first normative instruction on the control of deforestation issued by the new administration that assumed the Ministry of the Environment in 1999 (IN 002/99) greatly simplified the rules for the authorization of deforestation in the Amazon: for deforested areas up to 3 ha per year, in the case of family agriculture practised in small rural holdings (smaller than 150 ha), deforestation authorizations were practically made automatic, without the need of a previous technical visit to the area. This decision also responded to complaints of timber merchants and timber-industry of the region, who denounced “the extensive and increased extortion by government inspection and auditing agents in the region”. The new normative instruction had the support of the associations of agriculture producers and rural labourers of the region, as well as of the Timber Trade Federation and timber industry associations of Amazonia. The normative instruction was also widely supported and endorsed by all environmental NGOs, regional and national, which started to have a more participatory role and involvement in the newly-inaugurated administration of the Ministry of the Environment.

Box 4

IMAZON’s proposal for a control system of deforestation and forest exploitation in Amazonia

IMAZON has designed and proposed a system for the control of deforestation and forest exploitation in the Brazilian Amazon. One component of the system is technically based on FEMA’s system already in use. It includes, nevertheless, two additional elements:

(i) integration of environmental management of the state and of federal governments, with the proposition that the control functions be specifically divided between the institutions involved, on the basis that each rural property is controlled in all respects (i.e. deforestation/forest conversion and forest management) by only one institution.

(ii) “tracing back” of transported and exploited timber, based on the “OMNISAT” system. Authorized deforestation and authorized PMFS allow the concession of certain volumes of timber to be exploited and transported. But the timber could only be transported in vehicles that were equipped with a mobile communication terminal, constantly providing information on the location of the vehicle to the controlling institution (central). The driver would have to declare the origin, quantity and destination of the transported timber. In a central data bank the authorized volume (from management or from deforestation) would be automatically decreased/diminished by the quantity of transported timber.

The OMNISAT system is being used worldwide by 350,000 vehicles. In Brazil the system is operated by AutoTrac in more than 30,000 vehicles (private and public). The monthly cost is estimated at around US$ 110 for the leasing of the mobile equipment and US$ 50 for the data communication. Per year this means US$ 2000 per truck, or US$ 10-15 million for the whole timber transporting fleet of the Amazon. This means a cost of US$ 0.30 or US$ 0.50 per m3 of logs produced in the region (assuming a total of 30 million m3 of logs per year). These figures should be compared with the costs of SISPROF presented in Section 6 of this paper (especially with the costs/fees proposed/planned to be charged by IBAMA for the Stamps of Origin).

Other aspects to be considered are that (i) the OMNISAT system has been developed for business administration and management, not for enforcement and control to be exercised by the public sector; (ii) the system will have major implications for individual exploiters and companies, which are frequently small in size and scale of production, and which work/hire labour that have limited skills; (iii) the control of transportation, as proposed, does not seem to allow for an effective tracing back of processed/manufactured timber to its origin; (iv) finally, it must not be forgotten that, according to the Forest Code and the Law of Environmental Crimes (and to its regulatory Decree), timber and other forest products must be accompanied by documents (licences/permits) in all of their routes/trips.

As FoE/Brazil44 puts it, “the recognition that timber is the first short-term financial capital of an agrarian reform rural settler in Amazonia, and given the finding of a concrete impossibility of licensing and effectively following up his practices, has led the government to take an option for regulating in a simplified manner the raw material (timber) made available by deforestation and its injection into the commercial cycle”.

From this moment, the sawmills of Amazonia were suddenly presented with a cheap and simple alternative compared to what had certainly been made more expensive and complex given the progress and increase in the apparent inspection actions and control of timber exploitation and trade in the region, already referred to.

The sense and direction of these new rules would be further strengthened by another normative instruction (Normative Instruction no. 003/2002, quoted under, later to be issued by the Ministry of the Environment, in which the treatment and prerogatives given to small rural producers and settlers would be extended to landholdings or properties of up to 400 ha. DIREF of IBAMA, through a “Technical Note” to the Presidency of the Institute, warned about the consequences and risks of that new normative instruction.

Note the numbers involved with such a “regular” flow of timber in terms of the Amazonian timber production magnitudes: as we mentioned before, given a total official number of at least 300,000 settled families in the region, one could easily expect a “regular”, “simplified” potential injection of at least 14 million m3 (300,000 x 2.5ha/year x 20m3/ha) of logs into the timber market of Amazonia (a volume of 20m3/ha was fixed as the quantity of timber to be promptly conceded in a simplified deforestation authorization). This represents around 50 per cent of the regional annual production of 28-30 million m3 of timber (logs). It should be noted that this could also now represent the percentage of “legal”, “regularly authorized” timber in circulation in Amazonia, at least (there would still be the percentage of regular timber from apt or authorized PMFS).

In the face of the new rules, the new easily licensed supply of timber from agrarian reform settlements and other small rural properties in the Amazon would increase considerably and would end up promoting or leading to:

(a) The emergence of “intermediary agents” or middlemen (an extension of traditional loggers that operated in the region) that “channel” timber from settlements’ deforestation to sawmills, which contributes to reinforcing the reduction of vertical integration in the region’s timber industry, or to reinforcing the dependence of the industry on out-sourcing of timber or on the purchase of timber from third parties (a marked characteristic of this industry in Amazonia, as pointed out by authors such as Scholz, I.31).

As has been seen, one of the features of the “boom & bust” character of the timber industry in the region is precisely this unwillingness or aversion to invest in the formation of its own forest base. The resulting consequence has thus served to deepen/reinforce the industry’s dependence on deforestation timber and its associated logging practices.

(b) The need for that “intermediation” of channelling timber from settlements to sawmills to be further qualified. The simple and easy way of getting licensed volumes of timber from deforestation in these rural properties does not necessarily mean that these authorized volumes of timber will really and effectively be extracted from their corresponding deforestation sites and taken to the sawmills and processing units.

In fact, it is quite possible and likely that these easily obtained licences will serve (i) simply and only to cover up illegal/unauthorized (and not caught by inspection) extractions from any other sites, summing up the total authorized volume (in which case a “fee” would have to be paid to the settlers for the really non-existent, or non-extracted, timber from their properties), or (ii) to cover up, partially, illegal extractions from other sites, given that the extracted volume from the corresponding licensed sites would not have exhausted the total authorized volumes, in which case, also, a “fee”/price (implicitly or not, perhaps included in the price of the timber) might somehow have been negotiated between the “intermediary” and the settlers or small rural producers;

(c) The additional supply of easily authorized timber from deforested sites may have contributed to a possible reduction of “corruption fees/bribery costs” charged by corrupt officials in the region involved in fraudulent timber operations. It may also have contributed to reducing the covering-up of illegally obtaining timber through the use of “paper” or “ghost” PMFS, an alternative that, as we saw, used to be employed in the region but whose existence had been strongly combated through the extensive auditing and inspection operations undertaken by IBAMA.

(d) The new regime must have lowered transaction costs of getting timber for exploiters as well as for the timber industry;

(e) These new licensed volumes of timber thereby “injected” in the market of Amazonia will have been more than enough to keep the prices of timber sufficiently low (or to depress prices even further) and, therefore:

(i) will have surely contributed to worsening the competitiveness conditions of sustainably managed timber in the region, thereby making life more difficult for those that have engaged (or wanted to engage) in any sort of more environmentally sound activity, and in need of more economic profitability.

Prado, A.C.26 and Angelo, H. and Prado, A.C.39 have called attention to the fact that the two distinct sources of timber supply in Amazonia (deforestation x forest management) compete with each other, and that within present market conditions, timber coming from managed forests cannot compete with much cheaper timber coming from deforestation. The authors, therefore, suggest that to increase participation of forest management and of timber sustainably managed in the total supply of tropical timbers in the Amazon there is a need for measures that promote price differentiation between the two sources/types of timber, in favour of sustainably managed timber, of course, and not in favour of deforestation timber;

(ii) may not have represented, after all, a real advantage or gain to small rural producers or settlers (except, perhaps, in the very short term), given the outcome of lower market prices for timber;

(iii) served to deepen the ties and incentives that exist between timber exploitation and deforestation in the Amazon, reinforcing the significant propelling effect that the “boom & bust” timber industry exerts in the expansion of deforestation in the region; and

(iv) although the new regime might have eliminated or reduced the perverse action of “corrupt officials”, it did not eliminate, perhaps not even reduce, the unfair/fraudulent businesses or deals involved with timber exploitation and trade associated with deforestation in the Amazon.

In what could be seen as a pathetic attempt at public recognition and warning that these have been the likely results of the new rules imposed on authorizing deforestation in small rural properties and settlement areas in the Amazon, FoE/Brazil44 calls this new reality a “predatory legality”. Nice name for an old, widely known problem. Combating illegality is indisputably essential and necessary, and there is ample evidence that much progress has been achieved in this direction. Nevertheless, it is insufficient to revert the predatory and transient character of timber exploitation and the workings of the timber industry in the Amazon, whose incentives make them so deeply associated with deforestation2. On the other hand, to resort to measures, even if well-intentioned, in the name of simplification, for an alleged short-term alleviation of the harsh living conditions of the rural poor, may end up “turning the sorcery against the sorcerer”.

Be that as it may, there still are some complaints from parts of the timber industry, in terms of “operational difficulties” related to getting the licences, specifically from agrarian reform settlements in some areas of the Amazon, where the local INCRA offices have not been capable or willing to carry out the simple licensing due to difficulties in providing the necessary information on the allocated settlement lots or parcels.

These sectors of the industry have been using this type of argument to “justify” that they keep using PMFS in a fraudulent way, as a mere façade to cover up illegally obtained wood from deforestation (or from anywhere else), because of those operational obstructions. As if the difficulty (or facility) in obtaining licences for deforestation timber were the real determining factor in their using (or not) PMFS falsely or as a façade instrument. The ease in obtaining these licenses is certainly not what would make forest management an attractive or more competitive activity in the Amazon.

1 Most authors tend to single out, as the most important factor for the dynamism and expansion of industrial timber exploitation in the region, the public investments in the opening and building of roads, which made migration and deforestation possible in vast areas of Amazonia. Scholz, I., op.cit., also singles out the policy that from 1974 prohibited the exports of native timber in the form of logs from the region to foreign markets. The fiscal incentives of SUDAM for industrialization in Amazonia are not singled out as a very significant influence on the expansion of the timber industry itself, according to Scholz, op.cit. In fact, as she puts it, until the end of the 1980s SUDAM gave priority to the big investment projects. Out of the free trade zone of SUFRAMA, industrial projects of that scale only existed in mining and in cattle-raising, which absorbed the major part of fiscal incentives. The establishment of big cattle-ranching farms has always started with the purchase of vast areas of forested land with deforestation occurring in a part of the land, to prove/document possession and to obtain the fiscal benefits linked to the “improvements”. As these areas were often in remote places, there was no market for the felled trees/timber, which ended up being burnt. Small and medium investment projects that had no access to SUDAM incentives depended more on the sale of timber to finance the clear-cutting. This makes sense with the verification that, in the past, very high deforestation rates in Amazonia were not associated with a strong and dynamic presence and activity of the timber industry in the region.

2 No comments are provided here on other policies/measures already taken (or still to be taken) that, being more oriented to incentives instead of control, have resulted (or could result) in more efficient and effective achievement of the goals of forest conservation and sustainable forest management, as well as of reduction of deforestation in the Amazon. A detailed review and analysis of these other policies/measures may be found in Angelo, H. and Prado, A.C.39, op. cit.

Previous PageTop Of PageNext Page