Available online at www.sciencedirect.com
Forest Policy and Economics 10 (2008) 303 – 315 www.elsevier.com/locate/forpol
Property rights, land conflicts and deforestatio
n in the Eastern Amazon☆
Jose Antonio Puppim de Oliveira ?
Brazilian School of Public and Business Administration (EBAPE), Getulio Vargas Foundation (FGV), Brazil Land Laboratory (LaboraTe), University of Santiago de Compostela, Spain Received 12 March 2007; received in revised form 23 September 2007; accepted 28 November 2007
Abstract In the Brazilian Amazon, insecure property rights are among the main causes of land conflicts and deforestation. Through an in-depth empirical case study in Maranhao in the Eastern Amazon, this research analyzes how distorted agrarian, forest and environmental policies, laws and regulations originated insecure property rights not only over land, but also over timber, which allied to social and political factors, such as uneven distribution of land and strong organization of landless peasants, led to land conflicts and deforestation. This paper also shows that the causes of and the several actors involved in the deforestation of the Amazon were not independent, rather they were related and interact to each other. Compatibility between environmental goals and agrarian policies, regulations and laws are necessary to provide secure and clear property rights to allow a better enforcement of environmental regulations and to give actors incentives to avoid deforestation. ? 2007 Elsevier B.V. All rights reserved.
Keywords: Amazon; Deforestation; Agrarian conflicts; Property rights; Brazil; Maranhao
1. Introduction Changes in property rights regimes are perceived throughout the world. We have seen these great transformations which change enormously the property relations between individuals and societies over time, such as changes from feudalism to capitalism, from colonies to independent states, or from common to private property. In the recent policy arena, “hot issues” of changes in property rights are at the core of the development and forest policy debate, like the breakdown of the communist regimes in Eastern Europe (Williamson, 1994) or the dismantling of some traditional systems of common property (KabuboMariara, 2003; Ensminger and Rutten, 1991; May, 1990), or even allowing communities to trade communal rights (Engel and Palmer, 2006). Especially for the field of economics, and other social sciences related to political economy, this debate is crucial because property rights are assumed to be one of the bases of a well-functioning market economy. What are the causes and how
to explain changes in property rights and their consequences are the subject of significant work and debate in the New Institutional Economics (NIE) and other areas of social sciences that study institutions (Behera and Engel, 2006; North, 1990; Williamson, 1994; Alston et al., 1996; Demsetz, 1967; Libecap, 1989; Eggertsson, 1996). Particularly, there is a growing literature that focuses on the impact of property rights regimes and their changes on the natural environment (Ibarra and Hirakuri, 2007; Nelson et al., 2001; Marcouiller et al., 1996; Bromley, 1991). This paper examines the main causes of land invasions and agrarian conflicts in the region of Buriticupu, Maranhao State, in the Brazilian eastern Amazon frontier, and how, in some cases, these conflicts can become related to uncontrolled deforestation because of forest policy which did not favor forest management, forest conservation or preservation.1 Some of the literature has mentioned that insecure property rights over land are possible causes of land conflicts (Alston et al., 1995, 1996; Mueller et al., 1994) and deforestation (Casse et al., 2004; Binswanger, 1991). This research shows how land invasions
The author was at EBAPE/FGV during the reviewing process of this article. ? Tel.: +55 21 2559 5737; fax: +55 21 2559 5710. E-mail address: Joseantonio_puppimdeoliveira@yahoo.com.
1 In 2007, the Ministry of the Environment started a new initiative to give the concession of public forests to logging companies for controlled sustainable logging, but it is early to evaluate the results of this policy.
1389-9341/$ - see front matter ? 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.forpol.2007.11.008
J.A. Puppim de Oliveira / Forest Policy and Economics 10 (2008) 303–315
were caused not only by insecure property rights over land, but also by social and political factors and insecure property rights over timber, which result from poor enforcement of environmental regulations. In the case of Buriticupu the problems of insecure property rights over land and over timber, under certain social and political conditions, generated incentives to land invasions and, ultimately, deforestation. The paper also takes a different look at the most common explanations of deforestation in the Brazilian Amazon. The conventional wisdom in the literature generally assumes that the process of deforestation and environmental degradation in the region are caused by independent private actors—construction companies, loggers, farmers—stimulated by different government policies (Soares-Filho et al., 2004; Almeida and Campari, 1995; Schmink and Wood, 1992; Binswanger, 1991; Hemming, 1985; Moran, 1981; Mahar, 1989). Examples are the construction companies building large government-sponsored infrastructure projects, agro-industrialists implementing extensive agro-industrial projects subsidized by the government, timber loggers acting without adequate inspection by the environmental agencies or small producers from failed colonization projects practicing slashand-burn agriculture. In contrast with the common view, this research shows that the various private actors responsible for the deforestation in the Amazon may actually be interacting among themselves in their search for scarce resources such as land and timber. One of the main reasons for this interaction is the insecure conditions of the property rights over land and timber caused by distorted government policies, laws and regulations favoring unsustainable use of forests and land. As this paper presents, deforestation had only happened in this case as the result of the interaction among the different actors: landless people, landowners, logging industry and loggers. The research used the methodology of case study (Stake, 1995) by collecting quantitative and qualitative information together with open-ended interviews between 1996 and 2005. For this study, data was collected in Sao Luis (capital of Maranh?o State), Santa Luzia and Buriticupu. The distances of the studied areas varied from 0 to 40 km from the center of the town of Buriticupu. Around 50 open-ended interviews in the region were carried out with state officials, local government officials and politicians, community leaders, landowners, loggers, landless peasants and other relevant actors. The case describes the history of land occupations in the region between 1980 and 2000 approximately. 2. Changes in property rights Property rights can be interpreted as ‘the rights of an actor, group of actors or organizations to control certain resources' (adapted from Alchian, 1965), and are composed of a bundle of characteristics such as, exclusivity, inheritability, transferability, and enforcement mechanisms (Alchian and Demsetz, 1973). The system of property rights in a certain society could be defined as an institutional2 arrangement (Feder and Feeny, 1991). Property
2 North (1990) defines institutions as “the humanly devised constraints that shape human interactions,” or the formal or informal rules of the game in a society.
rights can reduce the uncertainties in human interaction, however, this does not imply that property rights regimes in place are economically efficient (North, 1990). As an institutional arrangement, property rights regimes have three dimensions of analysis: informal constraints, formal rules, and enforcement (North, 1990). Informal constraints are rules of behavior that people impose upon themselves to structure their relations with other members of society. In general, they are transmitted by tradition and are not formalized by written statements. In many developing countries informal rules are the main determinant of property rights over land and timber (Casse et al., 2004). Formal rules are written and formalized by constitution and laws, backed by the state. Enforcement is typically imperfect because there are costs and uncertainties for measuring performance of the rule to be enforced and because enforcement is performed by agents who may not have the capacity or interest in enforcing the rules. Therefore, to analyze changes in property rights, we have to look at the process of change in the formal rules, in informal constraints, and in the effectiveness and kinds of enforcement. One of the foci of the literature in diverse fields of social sciences is to explore some factors that are important to explain changes in property rights regimes. These factors include government intervention (Engel and Palmer, 2006), resource scarcity (Feder and Feeny, 1991), labor market (Nelson et al., 2001), prices, externalities (Demsetz, 1967), population density and migration (Otsuka et al, 2003; Feder and Feeny, 1991), political power (May, 1990; Ensminger and Rutten, 1991) and technology change (Libecap, 1989). For instance, Feder and Feeny (1991) describe how the definition of property rights in labor and land has changed over time in Thailand. In the nineteenth century, labor was scarce and land abundant, so there were very well defined property rights arrangements for labor, but not for land. As population density increased and land became scarce, property rights in land needed more definition to avoid conflicts. In Brazilian Amazon, changes in property rights regimes have transformed its agrarian structure (Otsuki et al, 2002). On the one hand, regimes of common property resources have been dismantled by agro-industrial interests and local elites, generating land degradation, emigration and conflicts (May, 1990; Anderson, 1990). On the other hand, peasant social movements occupied large land plots to accelerate agrarian reform. All these changes have had an effect on deforestation. This research shed light on how the process of deforestation occurred in Maranhao State in Eastern Amazon, one of the most conflictuous region in the Amazon, and its relation to changes in property rights. 3. Background of agrarian invasions and conflicts in Brazil Agrarian land invasions in Brazil are precipitated by organized landless people and their allies occupying public or large private properties, claiming that they are unproductive land. Therefore, those properties are susceptible to expropriation by the government for agrarian reform purposes, according to the law.3 The
3 The Brazilian constitution of 1988 includes the possibility of expropriation of unproductive latifundium for public and social interest, such as agrarian reform.
J.A. Puppim de Oliveira / Forest Policy and Economics 10 (2008) 303–315
Fig. 1. a—Map of Brazil (State of Maranh?o is the area in black) (Adapted from: Abril, 1994). b—Ecosystem of Maranh?o and region of study (municipality of Santa Luzia) (Adapted from: Arcangeli, 1987).
invasions are a kind of political pressure from organized landless groups on the government to accelerate the agrarian reform process. Invasions can lead to conflicts and violence between
landless peasants and landowners, when the latter resist the invasion. Agrarian conflicts have occurred frequently in Brazil since the 1960s. They are often the result of the struggle of
J.A. Puppim de Oliveira / Forest Policy and Economics 10 (2008) 303–315
landless peasants to get access to land rights against the resistance of large landowners to give up their land rights, in an unclear legal environment over who has the right over the land. 3.1. The region of study The area of study was the region around the town of Buriticupu, which was part of Santa Luzia municipality until 1997, in the state of Maranh?o in the Brazilian eastern Amazon. This region was chosen because it presented the highest number (21 in total) of land invasions in the state in the 1990s (Almeida, 1995) and related deforestation problems, which are considered priority issues for the state government of Maranh?o (SEMA, 1996), as well as other Amazon states. Indeed, the Maranh?o state had had several conflicts related to disputes over land and natural resources due to changes in property rights, large government projects and peasant struggles (May, 1990; Andrade, 1995). Buriticupu is located in the eastern part of the Legal Amazon,4 300 km southwest from the capital of the state, S?o Luis (see maps in Fig. 1a and b). The municipality of Santa Luzia had a population of 116,195 inhabitants, 85% living in the rural areas in 1991 (IPES, 1993). The urban center of Buriticupu had an estimated population of 15,000 inhabitants in 1996. In the 2001 Census, the municipality of Buriticupu had 50,839 inhabitants, 51,2% of them living in the town (IBGE, 2001). In the 1960s, the municipality was almost totally covered by the tropical forest, and its inhabitants were basically indigenous people and small farmers who practiced slash-and-burn agriculture, and a few cattle ranchers (Arcangeli, 1987). In the 1970s, the state government built a major road through the region and attracted many big agro pastoral projects with subsidies. These factors brought big changes, such as high population growth due to migration as well as unequal land ownership, which led to many social problems such as generalized poverty, land invasions and conflicts, and several critical environmental problems such as deforestation (SEMA, 1996) and land degradation due to unsustainable agricultural practices. 4. Conditions under which invasions took place Some of the literature suggest that the problem of insecure property rights over land in the Amazon frontier is one of the main causes of land conflicts (Alston et al., 1995, 1996; Mueller et al., 1994). The arguments is that the lack of effective land titling and secure tenure over land can spur land conflicts when some government action or policy takes place, such government subsidies for agro pastoral projects or infrastructure building,
pushing land prices up. This increase in land prices gives people incentives to try to obtain the land title, which in general is very difficult, because of the government agencies' lack of organizational capacity5 for a large increase in the demand for titling. This environment of insecure rights and increases in land prices lead people even to use violence in the fight for land rights. Infrastructure also creates access to exploit timber in more remote areas and spur the development of the timber industry (Soares-Filho, 2004; Stone, 1998). Moreover, common property resources used by peasants with no formal land registration were targeted to expropriation by large agro-industrial groups or local elites (May, 1990; Andrade, 1995). This process can increase land concentration, spur deforestation and augment social tension among peasantry, agroindustrial interests and landowners. An environment like the one described above was in place in Buriticupu in 1980. In the discussion, besides the factors that led to land conflicts described in the literature, this research adds, as accentuating factors in driving land conflicts and deforestation, the presence of insecure property rights over timber and its market value, as well social and political factors, such as uneven distribution of land and landless organization. On the other hand, in some situations, the same factors can also lead to interaction between the landless and settlers with local loggers and landowners, in order to exploit timber. In sum, this article points that the occurrence of a high number of land invasions and agrarian conflicts in Buriticupu was mainly related to the existence of four determinant factors: a) Agrarian conditions: the existence of vast, apparently unproductive land properties; b) Social factors: a large number of landless rural workers; distorted distribution of land resources; c) Political factors: democratization and the existence of a large momentum of social organization among workers and landless; and, d) Institutional factors (property rights): the existence of unclear and insecure property rights over land and timber, driven by conflicts between government agencies, by unclear laws or regulations, or by lack of enforcement. In the next sections these factors are described in more details. 4.1. Agrarian conditions: Amazonian colonization and the creation of large properties The recent interest in the colonization of the Amazon Region started in the 1960s during the military regime, which took power in 1964 (Almeida, 1995; Schmink and Wood, 1992; Mahar, 1989). The main concerns of the military government were the protection of the vast and under populated territory against the occupation by neighboring countries and supposed movements
4 There are two terms to refer geographically to the Amazon Region in Brazil. Legal Amazon is the term used for regional planning and policy (around 5.0 million km2). It comprises seven states (Acre, Amapá, Amazonas, Mato Grosso, Pará, Rond?nia and Roraima) and parts of another two (Maranh?o and Tocantins). The Amazon Forest Region is where the original vegetation was tropical rain forest or areas of transition between rain forest and other kinds of ecosystems (3.5 million km2 according to Ministerio do Meio Ambiente, dos Recursos Hídricos e da Amaz?nia Legal (1995)).
5 Including lack of human resources (quantity and quality), equipment and financial resources.
J.A. Puppim de Oliveira / Forest Policy and Economics 10 (2008) 303–315 Table 1 Tenure conditions in Santa Luzia Municipality in 1985 Tenure condition Owner (title holder) Renter or sharecropper (a) Squatter (b) Peasants holding no title (a + b) Total Number of families 4701 4443 7223 11,666 16,367
toward the internationalization of the Amazon. The federal government started implementing several plans aimed at fostering the occupation of the region, such as road building, telecommunication improvements, colonization projects and fiscal incentives for industrial and agricultural activities, especially large ones. The Amazonian Development Agency (SUDAM)6 and the Bank of the Amazon (BASA)7 were created to coordinate the process of development in the region. All these efforts to occupy the Amazon attracted millions of migrants into the region during the 1960s, 1970s, 1980s and 1990s (Almeida, 1995). The region of study (Buriticupu) followed the same path of development described above. In the 1970s, the state government built a major road through the municipality of Santa Luzia/ Buriticupu (road BR-222, between A?ail?ndia and Santa Inês). This road connected the region to the main trunk roads and regional markets. In the beginning of 1980s, a huge mining project, called Carajas Iron Project, started in the nearby state of Para. The mining company CVRD—Vale do Rio Doce (at that time a state company, later privatized) decided to build the infrastructure to exploit a large iron ore deposit in the Carajas mines. This project was within the federal government's larger Program Great Carajas for the development of the mining opportunities in the region, which is also rich in other minerals like bauxite and gold. CVRD transported the iron ore to the port of Sao Luis through a railroad that crossed the Buriticupu region. The Carajas Iron Project had an influence on the development of the region. Some authors, indeed, claim that Carajas was one of the main causes of the widespread deforestation in the Eastern Amazon (Andrade, 1995). During the same period, the State Agency for Land Colonization (COMARCO, which later became Maranh?o Land Institute-ITERMA8) created one of the most ambitious plans of colonization in the Maranh?o's Amazon. COMARCO incorporated 2.1 million hectares (ha) of state government land in the colonization plan in Santa Luzia/Buriticupu region. This plan aimed at distributing the land for three purposes: initiating a large colonization project for small farming for 10,000 families (315,000 ha), organizing approximately 3000 occupations already in existence (600,000 ha) and installing several agro pastoral development projects based on large properties (1185 million ha). Each of these projects had an average of 20,000 ha in area and received subsidized credit from the government (Luna, 1985). The plan did not yield the expected results. Only 1035 families were settled, and only 300 occupants had their land organized and registered in the end of 1980s. Concerning the
Percent of the total 28.7% 27.1% 44.2% 71.3% 100%
Source: IBGE, 1985—Agro pastoral Census of Maranh?o (pp. 196).
big development projects, COMARCO attracted 61 companies. Thirty-five of them still existed in 1982, occupying 700,000 ha (equivalent to 56% of the area of Santa Luzia municipality). Many of these projects closed down once the subsidized credit and incentives dried up in the 1980s.9 The failure of the settlements left an environment of poverty, landlessness and concentration of subsidized unproductive property in the region. 4.2. Social factors: population growth, landless peasants and distribution of land COMARCO's plan attracted thousands of migrants to the region in the 1970s and 1980s. As the state government advertised the colonization projects extensively, many migrants came hoping to get a piece of land in the colonization projects for small farmers. Others were attracted by possible job opportunities in some of the big agro pastoral projects, though in reality they were not offered in large numbers. Santa Luzia's population soared during this time (Buriticupu was part of Santa Luzia at that time). It doubled between 1970 and 1980, from 47,714 to 94,210 inhabitants. The population growth of Santa Luzia in the 1970s and 1980s is even more spectacular when it is compared to the growth in other regions, such as Maranh?o and Brazil. In the 1980s, the pace of population growth slowed down, but it was still two times rate of the state population growth between 1970 and 1991: the annual growth was 4.3% for Santa Luzia, compared to 2.4% for Maranh?o as a whole, and 2.1% for Brazil (IPES, 1993). The official colonization project for small farmers settled only a relatively small number of migrants. Thus, most of the incoming inhabitants pursued alternative routes for establishing themselves in the region, such as temporary labor for the agro pastoral projects, subsistence agriculture on public or private land (occupation or rent), and migration to the urban areas along road BR-222, especially the towns of Santa Luzia and Buriticupu.
6 Superintendência de Desenvolvimento da Amaz?nia (SUDAM) was created in 1966 and shut down by the Federal government in 2001 after evidences of widespread corruption practices and poor effective results in the development of the region. Recently, there are initiatives to reopen it again. 7 Banco da Amazonia (BASA) is the main supplier of long term loans in the Amazon region. It was created in 1942 as Credit Bank for Latex. 8 COMARCO (Companhia de Coloniza??o do Maranh?o, or Maranh?o Colonization Company). ITERMA (Instituto de Terras do Maranh?o, or Maranh?o Land Institute).
9 In 1979, SUDAM decided to disapprove livestock projects in the forested area, owing to an environmental outcry and a parliamentary investigation. Also, the volume of rural credit decreased after 1980, to tighten the balance of the Brazilian economy after the 1979 oil crisis. The subsidy for rural credit was eliminated in 1987 (Mahar, 1989).
J.A. Puppim de Oliveira / Forest Policy and Economics 10 (2008) 303–315 Table 2 Land distribution by category of size in Santa Luzia Municipality in 1985 Plots category Plots with more than 2000 ha Plots with less than 50 ha Number of land holders 52 14,250 Total area (ha) 410,882 83,684
The high population growth in the 1970s and 1980s distorted the distribution of land resources in the region, due particularly to extensive migration, and the attraction of people and companies interested in buying large properties,. Although the government distributed several titles to peasants in the settlements for small farmers, most of the families in the region could not obtain a land title. These skewed tenure conditions are illustrated in Table 1. In 1985, 71.3% of the families living in rural areas held no title. They were living on squatter plots, or renting or sharecropping a parcel of land from landowners. Besides the tremendous migration, other factors also helped to increase the number of workers looking for land. Some of these factors are the abandonment of plots in the agrarian reform settlements already established in the region,10 high birth rates, and changes in the traditional production relations on the farms.11 Other evidence for the distorted distribution of land resources in the region is observed in the division of the total area of land plots, titled or not, by categories (Table 2). In 1985, 52 persons with plots over 2000 ha held almost five times more land than the 14,250 persons whose land plots did not exceed 50 ha. This uneven distribution of land accentuated the social tension between have and have-nots. The landowners controlled the politics and were organized to protect their properties. The landless peasants had to rent land to produce or lived on the side of the road. Several groups organized landless people to occupy land. 4.3. Political factors: democratization and increase in the number of rural labor unions12 The federal and state governments allowed and gave incentives to the expansion of the rural labor unions (RLUs), and did this even during the military dictatorship (1964–1985). In the beginning, the planned function of the RLUs would be the management of social security (previdência social), and possibly later, the RLUs would serve as centers for medical and social assistance (Almeida, 1989). However, many of the RLUs also became involved in the organization of landless rural workers in their claims for land, especially after the democratization of the country in 1985. Not rarely, these RLUs were linked to left-wing parties and other movements of labor organization in the rural areas, such as Catholic church
Source: IBGE, 1985—Agro pastoral Census of Maranh?o, (pp. 252–225).
organizations13 and the Movement of the Landless Workers (MST).14 The number of rural labor unions (RLUs) in Maranh?o increased considerably between the 1970s and 1990s. In 1972, when the Federation of Rural Labor Unions of Maranh?o was founded, only 12 municipalities in the state had RLUs. A year later, there were 87 RLUs, and in 1981 this number had increased to 128 (Almeida, 1989). In 1995, every municipality in Maranh?o had established RLUs (136). Also, Maranh?o had a relative large number of RLUs, when compared to other states in the Amazon, in the 1980s before the beginning of the wave of land invasions (Table 3). Buriticupu had a representative to the RLU, whose headquarters was in the city of Santa Luzia. This representative organized workers in rural communities and mediated the acquisition of land for the landless workers. Besides the RLU, there was a strong rural worker organization organized by the opposition block to the directive of the RLU, which was linked to the Workers' Party (PT).15 Rural workers and landless people were highly organized in Maranh?o, particularly in Buriticupu region. This organization was the catalyst for the claim of peasants for obtaining their own land through changes in property rights in land invasions and agrarian reform movements. 4.4. Institutional factors: insecure property rights over land and timber 4.4.1. Contradiction between government agencies The National Institute for Colonization and Agrarian Reform (INCRA)16 and Brazilian Institute for the Environment and
10 Many settlers abandon their plots because of the lack of infrastructure (water, roads etc.) and social structure (schools, doctors, etc.), environmental and agricultural unsustainability (erosion, soil degradation or infertility), or just to search for more fertile or convenient land (many times selling the plot). 11 Sharecropping and renting land were very usual in the past. The increasing introduction of extensive cattle ranching in the region and the risk of land claims by peasants changed the traditional production relations on the farm. Many landowners preferred to end the sharecropping or rent contracts (Trov?o, 1989; Luna, 1985). 12 Rural Labor Unions in Brazil are organized at the municipal level. Municipalities also have representatives at the community level. At the state level, there is a Federation of Rural Labor Unions, and at the national level, there is a Confederation of Rural Labor Unions.
13 The Catholic church has several organizations working actively in rural community organization, such as the Pastoral Commission of Land (CPT) and Ecclesiastic Movement (MEB). 14 Movimento dos Trabalhadores Rurais Sem Terra (MST) is a national movement for organization of landless workers. It puts pressure on the political and judicial arena for agrarian reform. MST mainly acts in the organization, coordination and support of land invasions throughout the country. 15 PT, Partido dos Trabalhadores. 16 Instituto National de Coloniza??o e Reforma Agrária (INCRA) is a federal agency in charge of coordinating the whole agrarian reform process. Its activities are very diverse such as the expropriation of land, demarcation of plots, organization of settlers, provision of infrastructure and credit, and emission of land titles in settlements and colonization areas on federal land. INCRA activities are coordinated at state level and it has several regional offices all over the state.
J.A. Puppim de Oliveira / Forest Policy and Economics 10 (2008) 303–315 Table 3 Number of rural labor unions (RLUs) in the Amazon States in 1981 State Number of rural labor unions (RLU) 128 72 66 27 18 10 Number of municipalities in the state 136 105 211 95 62 12 Percentage of municipalities with Rural Labor Unions 94% 69% 31% 28% 29% 83%
Maranh?o Pará Goiás a Mato Grosso Amazonas Acre
Source: Almeida, 1989 and Larousse, 1991. a Tocantins State was part of Goiás State until 1988.
Renewable Resources (IBAMA)17 had had conflicting policies concerning land use in the Amazon, which could lead landowners to confusion about their rights over land and timber. One of the most polemical controversies between the two agencies was long-standing: IBAMA's policy for preservation of forest versus INCRA's criteria for considering a farm productive. Since the 1960s, when the national and international concerns about Amazon deforestation started, IBAMA had been in charge of enforcing a law that established that all properties in the Amazon had to leave 50% of the property as a forest reserve.18 Also, all developments in these properties or timber exploitation projects needed IBAMA's authorization for deforestation. On the other hand, INCRA used a national policy for expropriation for agrarian reform that was contradictory to IBAMA's policy for forest protection in the Amazon during many years. INCRA considered forested land as unproductive land in its calculation of the productivity of the land in the expropriation process. Therefore, the 50% of the property for forest reserve, as environmental regulation determined, was considered unproductive land for INCRA. This contradiction could make many landowners in the region uncertain about how to proceed for land development in their properties. If they developed more than 50%, they would be breaking an environmental regulation (and be susceptible to fines); if they developed less than 50%, they could possibly be susceptible to agrarian reform; an unsolvable dilemma. In the field research, two landowners reported that they were very confused about INCRA and IBAMA policies regarding land development, though they could not explain exactly what the agencies' policies were. One of them said in 1996: “I know that INCRA can expropriate unproductive land, though I do not
know the criteria…I have heard that people from IBAMA can arrest me if I deforest much of my property at once, but I do not know how they would know if I did this. They have never come here…” Another contradiction between the two agencies was the interpretation of the timber exploitation projects. IBAMA conceded to land owners authorization for timber exploitation in the Amazon region, which was subject to regulations concerning the area to be exploited each year, the area for reserve (a minimum 50% of the total area), the area for permanent reserve (riversheds, steep terrain, etc.) and the area to be left fallow or reforested. These projects, once approved, were subject to enforcement and evaluations. Although IBAMA and project owners would consider an area under exploitation of timber a productive activity, it seemed that INCRA (and land invaders) had not. INCRA had expropriated many timber projects in the region, considering them unproductive farms. A good example of this is the farm CIKEL in Buriticupu, which was invaded in May of 1996 and then expropriated, even though the owners of the property, who had a big sawmill in the city of A?ailandia, had a timber project approved by IBAMA operating according to the established official norms. 4.4.2. Property rights over land Several factors accounted for the confusion over land property rights in the region. The government policy that promoted settlements in the Amazon attracted many people who hoped to get a piece of titled land. However, the capacity of the government agencies (ITERMA and INCRA) to cope with the distribution and registration of land fell well behind the demand (Mueller et al., 1994). As a consequence, there were many rural families without land title. These families rented plots or occupied unclaimed government and private land. Many of these squatters occupied land that someone had already titled or had acquired the title after the occupation. Often the title holders managed to expel the squatters from the land, either negotiating or using violence or intimidation.19 The expelled squatters moved to other pieces of unclaimed land or became renters. Some of them ended up moving to urban areas. Even though there was a process for obtaining secure rights or title to unclaimed land through INCRA, this process was very slow and costly for the peasants, if it was done individually. Land claimants needed to gather several documents and travel several times to the closest INCRA agency (which in the case of Buriticupu was more than 100 km and at least 3 h away in the 1990s). Thus, the process of expulsion and the failure to secure tenure over land with title left many peasants searching for land and struggling to survive. Peasants could more easily acquire secure rights to land if they settle in one of INCRA's agrarian reform settlements. INCRA seemed to have a known policy of regularizing land
17 Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis (IBAMA) is a federal agency in charge of enforcing federal environmental laws and regulations. IBAMA's activities are coordinated at the state level, and it has several regional offices all over the state. By IBAMA, it means also the previous agencies that were accomplishing its role, as the Brazilian Institute of Forest Development (IBDF). 18 Law No. 4771 of September 15, 1965; article 16. This regulation changed in the 1990s. The area for forest reserve increased to 80% of the property in 1996 (Provisory Measure MP 1.511/96). There are some exceptions such as small holdings. However, this had a minimum of enforcement.
19 Sometimes the landowners paid the squatters to leave the land, or landowners used their private militias to threaten squatters or, using a judicial mandate, asked the police to expel the squatters.
J.A. Puppim de Oliveira / Forest Policy and Economics 10 (2008) 303–315
rights over settlements or claimant groups more efficiently than over individual plots. Having a INCRA land permit assured the secure right to use the land,20 which peasants sold frequently (there was a quite dynamic informal market for this kind of right). Moreover, INCRA had given priority to its scarce organizational resources to pursue agrarian reform in disputed areas, owing to political pressure to solve agrarian conflicts. In the region of study, most INCRA settlements (more than 20) were the result of land invasions, many involving violent conflicts. As a result, landless (or ‘titleless’) peasants perceived that land invasion was possibly the most accessible way of obtaining a secure right over a plot of land. Holders of land titles could also be subjected to losing their rights over land. Big landowners were susceptible to having their land expropriated for agrarian reform if INCRA considered the land unproductive.21 In this respect, the Brazilian agrarian reform legislation was very debatable and dubious. The Brazilian constitution stated that “land that is not accomplishing its social function is susceptible to agrarian reform,” meaning that to avoid agrarian reform, a property must be productive.22 But the concept of productivity was subjective, and it was not defined clearly. As a result, INCRA's decision whether or not a farm was productive was sometimes controversial. Some landowners considered farms that were invaded productive, but they were expropriated by INCRA anyway to avoid more social unrest and conflicts. In the field research, two landowners who had their land expropriated affirmed that their lands were productive when they were invaded by landless peasants (they also mentioned other supposedly “productive” properties that they said had had the same fate). They considered the expropriation more an action driven by political pressure, as one of them stated: “…the federal government does not want problems over its political reputation with the land invasions and conflicts. Because now the landless can get attention from the media with their invasions, INCRA expropriates any land invaded by them just to avoid more problems. INCRA officials do not care if the land is productive or not…”23 Both landless peasants and big landowners had limited de facto property rights over their plots. If we analyze the land tenure in terms of a bundle of property rights held by the different actors, we would reach the following conclusions: i) squatters had the right to use the land only for an uncertain period. When the piece of land was claimed by the title holder, sometimes they had the right to sell or be compensated for the
improvements they had made on the land. Squatters may acquire further rights if they registered the title through a slow and costly process.24 ii) Sharecroppers and renters had even more limited rights. They had the right only to use the land, paying a rent to the landowner through a formal or informal contract.25 They did not have the right to acquire the title and could be expelled anytime. iii) Settlers in agrarian reform plots had relatively secure titles, but they had to stay on the plot many years to have permanent secure titles, and most of the times they could not choose the plot (taking the risk to get an unproductive plot). iv) Landowners had the widest bundle of rights. They could use, sell and rent their land. However, if their plots were invaded these rights could be lost, though they were compensated.26 With the increasing number of successful27 land invasions, the uncertainties the landowners faced about their rights had increased. In sum, property rights over land in the region were not firmly secure, regardless of who holds the rights. This insecurity seems to be originated by several factors, such as INCRA's lack of organizational capacity, excessive bureaucratization of the process of titling and lack of clarity in the agrarian reform legislation. Moreover, some social factors accentuated the insecurity over land rights, such as high population growth, the high number of landless people, and the existence of a number of large properties. 4.4.3. Property rights over timber Property rights over timber were attached to the control over the land. These rights could be important to allocate labor and create incentives to different activities in the property (Otsuka et al, 2003). Since the control over land in the region was insecure, property rights over exploitation of timber were insecure as well. De jure property rights to exploit timber in the Amazon region belonged mostly to IBAMA (Brazilian Institute of Environment and Renewable Resources). Exploitation of timber or clearance of forest for agro pastoral activities on privately owned land needed IBAMA's legal authorization. According to IBAMA's regulations, all properties in the Amazon had to keep part of the property as reserve. However, IBAMA had only around 80 field inspectors28 in the Amazon to control an area of more than 5 million km2, with a large number of plots to check. Despite IBAMA's efforts to curb and control
20 The final title comes only after the emancipation of the settlement, which can take several years. 21 INCRA used the its Normative Instruction IN 08 of 1993 to analyze the Degree of Eficiency in Explotation (Grau de Eficiência de Explora??o or GEE) in order to decided wheter the farm can be expropriated. But these criteria were often disputed in court by landowners, and frequently they won the cause (getting more compensation for their land). 22 Brazilian Constitution, Title VII, Chapter III (Brasil, 1996). 23 However, in 2000, INCRA adopted a policy of not expropriating land while landless groups were occupying it.
24 The incentives to register the title can be increased, for example, when the value of the land increases (Alston et al., 1996). 25 Although sharecroppers had quite stable rights to use the land in the past, this changed because of the increase in cattle ranching activity and the risk landowners faced in having their land claimed by the landless. 26 The federal government used Titles of Agrarian Debt (TDAs) to compensate landowners for the land and cash to compensate for the improvements. The barrier to accepting this compensation was the lack of credit of the TDAs and the slow process of compensation (this process could last over a decade). 27 Successful land invasions are those which achieve their aims: obtaining land for the invaders, and making INCRA expropriate the land. 28 In 1996, during the time of land invasions in Buriticupu.
J.A. Puppim de Oliveira / Forest Policy and Economics 10 (2008) 303–315
deforestation in the Amazon, enforcement of the regulations was very feeble and impracticable.29 Moreover, corruption is a common practice in environmental enforcement. For example, 25% of IBAMA's employees in the state of Rio de Janeiro were arrested in 2006 (O Globo, 2006). It has been a critical problem in IBAMA over time, especially in the Amazon (ENS, 2006, Ecoagencia, 2007). With such weak enforcement, de facto property rights for timber exploitation mostly belonged to those who had control over the land where the trees were, and informal logging rules prevail. This happens even in small countries with more established forest policy, such as Costa Rica (Ibarra and Hirakuri, 2007). Landowners could exploit or sell the timber potential of their properties without control. Squatters and settlers could sell the timber on their plot. The areas that looked unclaimed or were public land often became open-access for timber, limited only by their accessibility. As result, loggers also became interested in the land disputes, as it is described below. Another factor encouraging the growth in the supply of timber was the existence of a growing demand for timber. In the early stages of settlement in Buriticupu, there was no market for the commercialization of timber. Most of the timber from the clearance of land was burned or used in the construction of houses. As timber companies started to appear in the region in 1980s, a growing demand for timber was created, and its value rose. The number of sawmills in Buriticupu increased from two at the beginning of the 1980s to 42 at the beginning of the 1990s. Moreover, the improvement of the established roads and the opening of access roads by loggers helped to diminish the accessibility constraints. A market for timber appeared. Those who had the property rights for the exploitation of timber could sell or use them. Since these rights were attached to the control over land, current control over land was de facto control over the exploitation of timber. In a region where the control over land was very vulnerable, these rights were susceptible to losses, either by expulsion in the case of squatters' land or by invasions in the case of landowners' land (a landowner is not compensated for timber if his or her land is expropriated). Also, because poverty was generalized, the discount rate for the settlers was very high. This induced them to sell the rights of exploitation very quickly and cheaply to whoever wanted to pay. Therefore, the insecurity over timber rights, the existence of a growing market for timber and poor enforcement of forest and environmental regulation gave people in the region incentives to quickly exploit the timber resources under their control.
5. Land invasions in Buriticupu and their relation to deforestation This section describes how the factors explained above have worked together to create an institutional and social environment that fostered invasions and deforestation in Buriticupu, through the interaction between the various actors: landless peasants, loggers, sawmill owners and landowners. 5.1. Landless organization in Buriticupu In the mid-1980s, as explained before, the number of landless peasants was very high in Buriticupu/Santa Luzia. Many of them were concentrated in the small villages or towns, along the main road of the region (BR-222). Buriticupu, as the headquarters of the COMARCO's colonization projects and the largest town in the region, contained many of these landless workers. During that time, the Brazilian military dictatorship was coming to an end,30 and democratization was advancing at a fast pace. In Buriticupu, few peasant leaders appeared and started to organize rural workers and landless people. Local organizations for this purpose had become stronger. These organizations included rural labor unions and organizations linked to the Catholic church and to left-wing parties. At the beginning of 1985, the former Brazilian President José Sarney31 opened the discussions for implementation of a supposed national plan to speed up agrarian reform and “invited” rural workers and peasants to help him.32 Local leaders used the President's call as an incentive to engage workers to organize themselves for the elaboration of a local agrarian reform project, though they were completely skeptical about the chances of implementation of this project. The rural workers were divided into 120 groups of 20 families each for the discussion of the lines of the project. The final project mentioned the name of 38 big properties in the region that the workers thought were unproductive, and therefore should be expropriated by INCRA. At the end, they submitted the project to the President and gave him a deadline33 to evaluate its feasibility. After the workers submitted the project, the leaders of the movement scheduled general meetings every 3 months to evaluate the government action regarding their project, keeping the workers' expectation about an unrealistic positive outcome very high. As the deadline approached, and as the leaders expected, there was no answer from the President or any other governmental authority. With the frustration of their expectation, landless people started sporadically invading some of the big properties mentioned at that project. Conflicts started because the landowners and managers of the properties knew about the rumors and prepared themselves for possible fights.
29 IBAMA pushed its effort to control commercialization and use of timber. IBAMA issued forms to loggers indicating quotas of timber for exploitation with information such as volume, destination and origin (ATPFs = Authorizations for Transportation of Forest Product or Autoriza??o para Transporte de Produto Florestal. In 2006, ATPF was substituted by DOF – Documento de Origem Florestal or Document of Forest Origination). Inspectors were delegated to periodically check trucks on the road and sawmills, to see whether they had those forms. But full enforcement was unrealistic most of the time because of the small numbers of inspectors for the size of the Amazon.
30 Brazilian military dictatorship started after a coup d'etat in 1964 and lasted until 1985. 31 Coincidentally, he was from Maranh?o. 32 Rural labor leaders say that this invitation was more for political than for practical purposes. 33 The deadline was December 25, 1985.
J.A. Puppim de Oliveira / Forest Policy and Economics 10 (2008) 303–315
All these movements culminated with the first big organized invasion of a farm called Capoema, and many others followed. The region became the most contentious area in the state, and possibly in the country. There were more than 20 occupations in 10 years (1983–93), as well as many other small conflicts (Almeida, 1995). 5.2. The appearance of the timber industry and its interaction with landless and settlers Meanwhile, the timber business gradually began to appear in the region and make connections with land invasions. As explained before, the property rights in the timber exploitation were held by those who had control over the land, which many times were not secure. Therefore, the “discount rate” of those who held the land was supposed to be very high. In the past, the constraint for the timber exploitation was the lack of a demand for it, which was limited by the accessibility of the region to the timber industry located at the nearby cities of A?ailandia and Imperatriz.34 Some sawmills were located in the Buriticupu region, but they had their own property and timber to exploit (sometimes regularized for timber exploitation by IBAMA), which was much higher than their sawing capacity. In the beginning, all the owners and managers of these sawmills were against the invasions, because some of them had big properties themselves, or they had strong links with the social elite of the region formed by landowners and farm managers. On the other hand, the leaders of the invasions and squatters themselves knew about the political position of the sawmills in Buriticupu. As a consequence, in the regular clearance of the squatter land and even in the first invasions, a small part of timber was used for building, but most of it was burned for preparing the land for planting, or was left unused. As the timber became scarce in the neighboring Imperatriz and A?ailandia and the access from this region to Buriticupu was improved, some madereiros35 started traveling to Buriticupu in search of timber. At the same time, squatters and settlers started to perceive timber as an opportunity to earn income. Also, they did not see the madereiros from other regions as a threat to them and their situation, because these madereiros, different from the traditional ones, were from outside the conflict region and had almost no connection to the “enemy” (the local elite: big local land holders). So, a fruitful relation between the growing timber industry and settlers/squatters began. As the number of occupations increased and the reputation of the organized social movements spread, different people from the same region or from other regions started joining the invasion movements. These people were not homogeneous. The different kinds of invaders included landless people looking for a piece of land for agriculture, local people interested in
obtaining a piece of land to sell or to have as an asset reserve and people interested in exploiting the timber potential of the big properties rather than land. This last group (this group is called “loggers” or “timber group”) was composed of madereiros, motoqueiros36 and small timber merchants looking for a nearby supply of cheaper timber for the A?ailandia and Imperatriz timber industries. Because most of the people in the timber group were new in the region and poor,37 it seems that they did not receive any opposition from the invasion movements. To the contrary, they were very welcomed because the philosophy of the movement leaders was “as more people join the movement, the better.” They thought that more people would make the movement stronger. Moreover, interaction links started occurring between loggers and the rest of the invaders. Loggers gave logistic support for the invasions such as transportation of people and food supply. Loggers made agreements with invaders to “buy” the logging concession to exploit their future properties (paying a modest price). The landless peasants, many of them only interested in having their own pieces of land, promptly accepted these agreements, because timber did not have any utility for them (aside from using it for fuel), and they needed cash and help to support themselves in the beginning stages of the settlement38 for building houses and subsistence until the first harvest. Loggers also helped settlers in many aspects of daily life. I noticed in my field research that, after invasion, loggers provided transportation, opened roads, did some works (such as building small dams or clearing land), brought water and even served as middlemen for commercialization of farmers' products. These activities were more positive externalities of the logging activity. Once the timber becomes scarce, the loggers stopped coming. As the process of establishing an INCRA agrarian reform settlement was very slow and the improvements uncertain, settlers established a quasi-dependent relation with loggers sometimes. So, this relationship satisfied the interests of both parties. As the number of invasions grew, many loggers moved to Buriticupu. They were involved in one invasion after another in search of timber. Moreover, some of them become leaders in the invasions. They also started to expand their action through the properties invaded before.39 Meanwhile, the establishment of small and medium-sized sawmills soared in Buriticupu, as mentioned before. Some of these sawmills started to move from other regions where timber was becoming scarce, such as Imperatriz, to Buriticupu. Others were just start-ups from people
34 A?ailandia is 150 km and Imperatriz 250 km south of Buriticupu. They were the centers of the timber industry in the 1970s and 1980s. 35 A madereiro is a “Lumberman living from the extraction and transportation of logs to timber merchants or sawmills.” (Ros-Tonen, 1993). Some madereiros are self-employed, while others work for timber companies.
Motoqueiro is a chainsaw operator. In general, the timber group people had only their means of work: a old falling-apart truck or a chainsaw. Many times they were landless. I met some people who actually lived in the trucks or in the forest in improvised tents and moved from place to place of work. 38 Once the invasions became agrarian reform settlements, INCRA started to give financial and infrastructure support for them. But the support was uncertain and can take many years. 39 Settlements remained forested for a long time, if no logging or extensive production activity is introduced. This was because settlers continued with the same kind of low land-use, slash-and-burn agriculture.
J.A. Puppim de Oliveira / Forest Policy and Economics 10 (2008) 303–315
in the region who saw a good opportunity for business. Also, some of the sawmills that opposed the invasions started buying timber from the loggers, who were mostly from outside the region and did not participate in the first invasions, and who therefore did not see the old sawmill owners as opposition. With the increasing number of sawmills and intermediaries in the timber business, the demand for timber increased sharply, and as explained before, the property rights for timber exploitation were attached to the control over land. So this structural change in the market for timber generated a new scarce resource for pressuring the land invasions: timber. 5.3. Impacts of the invasions on landowners' behavior For the big landholders in the region, the land invasions represented a risk of losing not only the power over land, but also the timber potential of their property. In the expropriation process after a land invasion, the loss of the land and its improvements were compensated by INCRA through an evaluation and an agreement with landowners. But the timber potential was not compensated. INCRA did not consider the forest reserve as an improvement subject to compensation. So, as IBAMA control was all but present, landowners had incentives to log the valuable trees as quickly as possible in a region under risk of invasion, as timber could be more valuable than the land sometimes. According to the interviews, typical value for a hectare of cleared land in the region valued between 100 and 200 Real (R$) (one Real was approximately $US 1 at the time of the research) and a typical market value for the hardwood found in one hectare of land could be sold for 1000 Reals (not including the logging costs). Moreover, as INCRA considered pasture land as eligible for compensation, landowners had a double incentive to deforest and create pasture land. They earned twice: from the timber and from the pasture. This process of change in the use of land from forest to pasture was further accelerated by the increasing number of conflicts in the region, which was proportional to the risk of having one's farm invaded. This process was noticed in two of the farms studied: Cikel (already invaded) and Frechal. The Cikel farm belonged to a big timber company. This company had permission from IBAMA to extract timber legally according to a contract. When small invasions and conflicts started in Cikel and the owners heard rumors about a possible future invasion of their land, they started to increase the rate of exploitation.40 When the farm was definitively invaded, a large quantity of fallen trees was found (a few thousands of cubic meters of timber), showing that the owners were trying to extract as much timber as possible before the invasion took place. In Frechal farm a similar process happened. Another consequence of the above-mentioned policy was the landowners' incentive to organize invasions on their own land. As the public financial incentives for the Amazon dried up,
many companies began to suffer from economic difficulties. Selling the land could be their salvation. But since land invasions spread out over the region, the price for land had fallen increasingly in the regional market, according to local brokers. This was logical: someone who knew the region would be unlikely to buy a big property there, since it could be the target of a future land invasion. Moreover, land prices in Brazil were falling all over the country due the control of inflation (Gazeta Mercantil, 1996).41 The solution that some big owners had found was to induce land invasions in their own plots through agreements with the leaders of the landless movement,42 and then to get compensation from INCRA. Although the compensation process was slow and the compensation for the land was paid using the devaluated Agrarian Reform Titles (TDAs), in general INCRA's compensation price per hectare was over market prices and improvements were paid shortly after the expropriation.43 Besides, TDAs had increased their values in the market because they could be used as money in the process of privatization of public firms, which the government had started to sell in the 1990s. However, the invasions were “planned” in such a way that the landowner had time to harvest all the accessible valuable trees and to “make” fake improvements in form of pastures to get more compensation. 6. Conclusions In the Brazilian Amazon, distorted governmental policies, laws and regulations generated insecure property rights over land and timber. In turn, these insecure rights, allied to social and political factors, spurred land invasions and conflicts, and resulted in environmental degradation, especially where there was a lack of enforcement of environmental laws. Moreover, the same policies could induce the main private actors in the process of deforestation (landless people, loggers and landowners) to interact between themselves, as in the Buriticupu region. The result of this research presented the distortions in some Brazilian policies for agrarian reform and environmental protection, such as lack of criteria for land expropriation, difficulties in the process of land titling, conflicting policies between INCRA and IBAMA and poor enforcement of environmental laws, such as limitation on forest clearing. These policies fostered an environment of insecure property rights whose consequences included (i) the lack of incentives for private investments on land improvements; (ii) rent dissipation in organizing invasions (in the case of the landless) and in protecting properties against invasion (in the case of the landowners); (iii) waste of public resources in conflict resolution and hurried (and sometimes unfair) land expropriations;
40 They used the clearing system called “corrent?o” (big-chain), which means two tractors connected by a huge-thick metal chain pushing the trees down.
41 As inflation was apparently controlled, investors reduced investments in real state as hedge against inflation, so land prices fell. 42 I found two cases in my field research: Frechal and Varig farms (already occupied). 43 Besides, the values for land can be easily contested in courts. Sometimes these values increase many-fold at the end of the process according to INCRA specialists.
J.A. Puppim de Oliveira / Forest Policy and Economics 10 (2008) 303–315
(iv) violence and social unrest in some cases; and (v) uncontrolled deforestation and land degradation. Compatibility between environmental goals, and forest and agrarian policies, regulations and laws are necessary to provide secure and clear property rights to allow a better enforcement of environmental regulations and to give actors incentives to avoid deforestation. In the case, traditional and informal land rights existed and were indisputable and consensual, land titling was a mechanism for registration of rights, allowing access to credit and formal land markets (Feder and Feeny, 1991) or even reduce conflicts and deforestation (Godoy et al., 1998). In these cases, land titling generally would not cause much social disruption. However, land title and credit can lead to an increase in deforestation in some models (Angelsen, 1999). In practice the agricultural intensification may result in lower rates of forest clearing in certain conditions, for example when agriculture drives labor demand out of deforestation activities (Nelson et al., 2001). Moreover land titling programs with redefinition and redistribution of rights can cause unintended effects at local level such as, social disruption or conflicts (Mueller et al., 1994). Redefinition of land rights may be needed where the traditional local system of land tenure changed, and in turn, these changes may be one of the causes of land degradation (Wachter, 1992). Also, redefinition of rights, in form of an extensive land reform, may be inevitable to redefine disputable or unfairly distributed land rights where they impede any efforts toward land conservation and control of deforestation (Otsuki et al, 2002). This seemed to be the situation in parts of the Brazilian Amazon, where in many cases, land rights and titles of big unproductive properties were obtained under unclear circumstances (bribing, cheating, violence, political favor etc.); and on the other side, millions of landless workers need land to survive. This unbalance in the rights led many landless people to move farther in the forest to look for a piece of unclaimed land, causing further deforestation, or occupied large plots of land asking for agrarian reform. Therefore redefinition of these rights with clear rules seems to be needed to control deforestation and land degradation, but this has to come with a much stronger environmental law enforcement regime in order to make clear property rights over timber. Recently, the government in its different levels have tried to coordinate development and conservation policies with promising results (Fearnside, 2003). Moreover, market based initiatives, like forest certification (especially FSC), are growing (Cashore et al., 2006), so there is hope for change. Acknowledgement I would like to thank the institutional support of the government of Maranhao and the research grant of the Department of Urban Studies and Planning of the Massachusetts Institute of Technology for the first field work. I would also thank the support of the Brazilian School of Public and Business Administration (EBAPE), Getulio Vargas Foundation (FGV), Brazil, for the last part of the field work. I appreciate the comments in the early versions of this research by Larry
Susskind, Judith Tendler, Monica Pinhanez, Rodrigo Serrano, Marcela Natalicchio, Judy Morrison, Ann Steffes and Tito Bianchi. However, the data and opinions are of my sole responsibility. References
Abril, 1994. Almanaque Abril. Abril, S?o Paulo. Alchian, A.A., 1965. Some economics of property rights. Il Politico 30. Alchian, A.A., Demsetz, H., 1973. The property rights paradigm. Journal of Economic History 33, 16–27. Almeida, Alfredo Wagner Berno, 1995. The state and land conflicts in Amazonia, 1964–88. In: Goodman, David, Hall, Anthony (Eds.), The Future of Amazonia. Macmillan Press, Hong Kong. Almeida, Alfredo Wagner Berno, 1989. O movimento sindical dos trabalhadores rurais e os conflitos agrários na Amaz?nia. Pará Agrário 6/7. Almeida, Anna Luíza O., Campari, Jo?o S., 1995. Sustainable Settlement in the Brazilian Amazon. The World Bank, Washington D.C. Alston, Lee J., Libecap, Gary D., Schneider, Robert, 1995. Property rights and preconditions for markets: the case of Amazon frontier. Journal of Institutional and Theoretical Economics 151 (1), 89–107. Alston, Lee J., Libecap, Gary D., Schneider, Robert, 1996. The determinants and impact of property rights: land titles on the Brazilian frontier. Journal of Law, Economics & Organization 12 (1), 25–61. Anderson, Anthony B., 1990. Deforestation in Amazonia: dynamics, causes and alternatives. In: Anderson, A.B. (Ed.), Alternatives to Deforestation. Columbia University Press, New York, USA. Andrade, Maristela P., 1995. Carajás: Desenvolvimento ou Destrui??o. CPT, S?o Luis, Brazil. Angelsen, Arild, 1999. Agricultural expansion and deforestation: modelling the impact of population, market forces and property rights. Journal of Development Economics 58 (1), 185–218. Arcangeli, Alberto, 1987. O Mito da Terra: Uma Análise da Coloniza??o da PréAmaz?nia Maranhense, UFMA, S?o Luís, Brazil. Behera, Bhagirath, Engel, Stefanie, 2006. Institutional analysis of evolution of joint forest management in India: a new institutional economics approach. Forest Policy and Economics 8 (4), 350–362. Binswanger, H., 1991. Brazilian policies that encourage deforestation in the Amazon. World Development 19 (7), 821–829. Brasil, 1996. Constitui??o Brasileira de 1988, 8th edition. Editora Atlas, S?o Paulo, Brazil. Bromley, Daniel W., 1991. Environment and Economy: Property Rights and Public Policy. Basil Blackwell Ltd, Oxford, England. Cashore, Benjamin, Gale, Fred, Meidinger, Errol, Newsom, Deanna, 2006. Forest certification in developing and transitioning countries. Environment 48 (9), 6–25. Casse, Thorkil, Milh?j, Anders, Ranaivoson, Socrate, Randriamanarivo, Jean Romuald, 2004. Causes of deforestation in southwestern Madagascar: what do we know? Forest Policy and Economics 6 (1), 33–48. Demsetz, H., 1967. Towards a theory of property rights. American Economic Review 57, 347–359. Ecoagencia, 2007. Opera??o Ananias desarticula quadrilha no Pará. www. ecoagencia.com.br, March 02. Eggertsson, T., 1996. A note on the economics of institutions. In: Alston, Lee J., Eggertsson, Thráinn, North, Douglas (Eds.), Empirical Studies in Institutional Change. Cambridge University Press, New York. Engel, Stefanie, Palmer, Charles, 2006. Who owns the right? The determinants of community benefits from logging in Indonesia. Forest Policy and Economics 8 (4), 434–446. ENS—Environmental News Service, 2006. Brazil Fells Massive Amazon Timber Fraud Ring. www.ens-newswire.com, June 6, 2005. Ensminger, Jean, Rutten, Andrew, 1991. The political economy of changing property rights: dismantling a pastoral commons. American Ethnologist 18 (4), 683–699. Feder, Gershon, Feeny, David, 1991. Land tenure and property rights: theory and implications for development policy. The World Bank Economic Review 5 (1), 135–153.
J.A. Puppim de Oliveira / Forest Policy and Economics 10 (2008) 303–315 Fearnside, P.M., 2003. Conservation policy in Brazilian Amazonia: understanding the dilemmas. World Development 31 (5), 757–779. Gazeta Mercantil, 1996. Cota??es da terra em queda livre. Gazeta Mercantil Newspaper, pp. A–11. December 19. Godoy, Ricardo, Jocobson, Marc, De Castro, Joel, Aliaga, Vianca, Romero, Julio, Davis, Allison, 1998. The role of tenure security and private time preference in neotropical deforestation. Land Economics 74 (2), 162–170. Change in the Amazon Basin. In: Hemming, John (Ed.), The Frontier after a Decade of Colonization, vol. II. Manchester University Press, Manchester, U.K. Ibarra, Enrique, Hirakuri, Sofia R., 2007. Institutional conflict and forest policy effectiveness: the case of the Costa Rican institutional reform. Forest Policy and Economics 9 (6), 591–601. IBGE, 1985. Censo Agropecuário do Maranh?o. IBGE, Rio de Janeiro. IBGE, 2001. Popula??o Estimada por Municípios. . Home-page: www.ibge.gov. br. August. IPES, 1993. Sinopse Estatística Municipal do Maranh?o, IPES, S?o Luis, Brazil. Kabubo-Mariara, Jane, 2003. The linkages between property rights, migration, and productivity: the case of Kajiado District, Kenya. Journal of Environment and Development Economics 8 (4), 621–636. Larousse, 1991. Brazil de A/Z, Larousse Cultural, S?o Paulo. Libecap, Gary, 1989. Distributional issues in contracting for property rights. Journal of Institutional and Theoretical Economics 145, 6–24. Luna, Celi M.R., 1985. A Terra Era Liberta: um Estudo da Luta dos Posseiros no Vale do Pindaré. UFMA, S?o Luís, Brazil. Mahar, Dennis J., 1989. Government Policies and Deforestation in Brazil's Amazon Region. The World Bank, Washington, D.C. Marcouiller, D., et al., 1996. Timber production factor shares by forest tenancy group. Land Economics 72 (3). May, Peter, 1990. Palmeiras em Chamas. S?o Luis, Brazil, EMAPA/FORD/ FINEP. Ministerio do Meio Ambiente, dos Recursos Hídricos e da Amaz?nia Legal, 1995. Os Ecosistemas Brasileiros e os Principais Macrovetores de Desenvolvimento. MMA, Brasília. Moran, Emilio F., 1981. Developing the Amazon. Indiana University Press, Bloomington, USA. Mueller, Bernardo, Alston, Lee J., Libecap, Gary D., Schneider, Robert, 1994. Land property rights and privatization in Brazil. The Quarterly Review of Economics and Finance 34 (Special Issue), 261–280.
Nelson, Gerald C., Harris, Virginia, Stone, Steven W., 2001. Deforestation, land use and property rights: empirical evidences from Darien, Panama. Land Economics 77 (2), 187–205. North, Douglass C., 1990. Institutions, Institutional Change, and Economic Performance. Cambridge University Press, Cambridge. O Globo, 2006. Opera??o da PF Prende ? do IBAMA no Rio. O Globo newspaper. August 08. Otsuka, Keijiro, Quisumbing, Agnes R., Payongayong, Ellen, Aidoo, J.B., 2003. Land tenure and the management of land and trees: the case of customary land tenure areas of Ghana. Journal of Environment and Development Economics 8 (1), 77–104 February. Otsuki, Tsunehiro, Hardie, Ian W., Reis, Eustáquio J., 2002. The implication of property rights for joint agriculture–timber productivity in the Brazilian Amazon. Journal of Environment and Development Economics 7 (02), 299–323. Ros-Tonen, M.A.F., 1993. Tropical Hardwood from the Brazilian Amazon: A Study of the Timber Industry in Western Pará. Verlag Breitenbach, Saarbrucken, Germany. Schmink, Marianne, Wood, Charles H., 1992. Contested Frontier in the Amazon. Columbia University Press, New York. SEMA—Secretaria de Meio Ambiente e Recursos Hídricos do Maranh?o, 1996. Plano Estadual Ambiental, Report. SEMA, Sao Luis, Brazil. Soares-Filho, B., Alencar, A., Nepstad, D., Cerqueira, G., Diaz, M.C.V., Rivero, S., Solorzano, L., e Voll, E., 2004. Simulating the response of land-cover changes to road paving and governance along a major Amazon highway: the Santarém–Cuiabá corridor. Global Change Biology 10. Stake, Robert E., 1995. The Art of Case Study Research. Sage Publications, Thousand Oaks. Stone, Steven W., 1998. Evolution of the timber industry along an aging frontier: the case of Paragominas (1990–1995). World Development 26 (3), 433–448 March. Trov?o, José Ribamar, 1989. “Ilha” Latifundiária na Amaz?nia Maranhense,. UFMA, S?o Luis. Wachter, Daniel, 1992. Land titling for land conservation in developing countries? Working Paper No. 1992-28, The World Bank Environmental Department, Policy and Research Division. The World Bank, Washington, D.C. Williamson, Oliver E., 1994. The institutions and governance of economic development reform, Proceedings of the World Bank Annual Conference on Development Economics. The World Bank, Washington, DC.