Brazil: land politics, poverty and rural development - E. Teofilo and D. Prado Garcia

Edson Teofilo, Director Projeto Crédito Fundiário e Combate à Pobreza Rural
(Fighting Rural Poverty and Land Credit Project)
Danilo Prado Garcia, Núcleo de Estudos Agrários e de Desenvolvimento Rural

This article puts into an historical context land reform issues in Brazil as related to sustainable economic development. Agrarian reform, rooted in the creation of social capital, provides a basis of growth within which we have framed this discussion. Although the economic and social costs of agrarian reforms are high and time negatively affects those beneficiaries who are not competitive because there is no regulated and accepted access to land (titling), it is necessary to incorporate in agrarian reform interventions the "new" approaches of rural development: regional growth (context) and economic incorporation of the poor (active).

HISTORICAL DESCRIPTIONS

One characteristic of Brazilian development has been its disregard for evidence that suggests that sustainable economic development is intimately connected with the issue of land distribution. This oversight is symptomatic of the conservative nature of the development process in Brazil. For many years, the question of agrarian reform, and specifically market-based land reform, has been presented in diverse forms and with varied intensity; and although the resultant policies have not always been coherent, rational dissention has been largely absent.

In Brazil, an archaic system of property (i.e. land) rights supports one of the world's most iniquitous and inefficient land distribution systems. However, the issues are substantively different today than they were at the end of the nineteenth century or even in the 1950s. Growth of the Brazilian economy in the last decade(s) begs the question of whether the agrarian structure was in fact an obstacle to the country's economic development. Enlargement of the domestic market^[1] and globalization may well have facilitated analysis that overlooked the fundamental question of capital accumulation based on property rights.

The question of food safety - another historical argument in favour of agrarian reform - has also been redefined through the process of economic integration and globalization: the correlation between food safety and self-sufficiency that guided some agricultural policies in the past has been replaced by a concept of self-reliance, which emphasizes the capacity to obtain food, rather than to produce it.

A number of factors seem to suggest that the agricultural sector has in fact been modernized: today, agribusiness is responsible for a significant percentage of the country's farming production; part of the latifundia farms have acquired an entrepreneurial character that has resulted in increased levels of productivity; the old system had been replaced by a modern one, the caveat being that even this modern system falls short of the standards reached by Organisation for Economic Co-operation and Development (OECD) countries; and finally, Brazil has become a producer and exporter of non-traditional agro-industrial products. Despite the negative effects of the "Eighties' crisis", the farming sector has been modernized and today is considered an efficient and competitive sector of the economy.

Despite this progress, gaps remain in its modernization that seem to have been facilitated by a closed trading regime, which artificially protected the sector, and resulted in concentration of wealth during periods of monetary instability. Thus, the productive transformation, far from alleviating rural problems, has facilitated their continuation and has even aggravated the situation.

The conservative modernization of the latifundia farms reinforced concentrated land ownership and the exclusionary character of the "farming development model"; in general, the old system was replaced by a system characterized by temporary employment, which in many regions offers no legal protection to the workers. In some areas, forced labour still exists in an obscure form similar to slavery, where child labour and deplorable conditions are the norm. Subsistence farming has been largely eliminated, leading to increased migration to urban areas. Many of the existing small farms today serve as homes, rather than as units of production; the rural population is increasingly migrating to large- and medium-sized urban centres that struggle to accommodate them.

Recently, the magnitude and visibility of the agrarian problem, in large part a continuation of the drawn-out crisis that has affected Brazilian agriculture since the end of the 1980s, has become even more evident as displaced rural populations without land or work search for other means of survival. It is in this context that violence and conflict in rural areas has grown, further necessitating societal intervention.

Different views on the agrarian question in Brazil

During the 1960s the question of agrarian reform in Brazil was a relevant issue in both political and academic circles. Rural land concentration, a heterogeneous production system, labour relations, the existence of precapitalist structures such as communes, and the poor quality of life for agricultural employees were the main subjects to be tackled. Among the classic interpreters of the origins of and solutions to the Brazilian question of land reform are Ignácio Rangel, Alberto Passos Guimarães and Caio Prado Jr. For Ignácio Rangel and Alberto Passos Guimarães, the question of land reform is a function of the difficulty inherent in attempting to adapt a capitalist system to rural areas characterized by archaic and often feudal agricultural structures inherited from the colonial system. Thus, they believe the solution requires overcoming the obstacles to the development of a market-based system. For Caio Prado Jr the question of land reform derives from the process of capitalist development (Kageyama, 1993).

For Ignácio Rangel (1962), the resolution of the land reform question requires the transformation from a self-sufficient agricultural model to a set of superior forms of organization based on free market ideals (private companies and individual entrepreneurs). The result of this transformation, however, was disequilibria and a subsequent crisis of overproduction and overpopulation. The excess labour in rural areas and urban peripheries could not be absorbed by the market economy. Therefore, Rangel's proposed solution was to reconstitute the natural economy on a small scale, with the objective of guaranteeing the subsistence of this excess population. This would be achieved through the creation of small family plots in both rural and peri-urban areas (Rangel, 1962, pp. 23 - 27).

Rangel discourages any attempt to change the agrarian structure through governmental land purchase. By contrast, the state must supply the land, using its own land or inducing the private sector to make land available. In contemporary Brazil, the land problem is essentially an economic rather than a legal problem. Intervention by the state as the buyer would distort price adjustment and cause other distortions in the economy, for example through the tax system, that would hinder structural change (Rangel, 1962, p. 26).

Alberto Passos Guimarães concentrates his arguments on the inheritance of the colonial latifundia farms, which he sees as having left a legacy of obstacles including archaic labour relations characterized by extra-economic coercion and personal dependence. He believes the need for agrarian reform emerges from the incompatibility between archaic agrarian structures and the capitalist system. His proposal to overcome those obstacles includes the elimination of the inefficient and "old-fashioned" latifundia system (Kageyama, 1993, pp. 7 - 8).

For Caio Prado Jr the question of land reform and its problems are a consequence of the imposition of a capitalism system, which has reduced the system's ability to interpret and address its internal problems. His approach subsequently becomes a Marxist analysis, which focuses on labour relations in Brazil's rural economy. For Prado, the question of land reform is embodied in the rural population's lack of material possessions and their lack of legal rights. It is not a result of the colonial feudal legacy, but rather of the existing capitalist structures that have characterized Brazil's economic development. The solution should therefore be predicated upon the available instruments and institutions, with all of their shortcomings (limited property rights, labour laws, minimum wage, etc.) and only in a second phase should there be an attempt to replace the system entirely (Kageyama, 1993, pp. 8 - 11).

It is important to point out that these authors' analyses of land reform are not analogous to analyses of ownership concentration. Although property rights and the historical forms of its occupation have been reflected in Brazil's agrarian problems, the expression of these problems is related to the population, either in terms of excessive structural unemployment (exceeding population, of Rangel), worker exploitation (Guimarães) or lack of legal protection, all of which perpetuate agricultural poverty (Prado).

The analysis of the agrarian question based on income from property does not take into account that, in a capitalist economy, land is legally defined as private property, and as such is subject to the price mechanism and market forces. Although "land" was not an economic good as such, it became one during the dramatic nineteenth century transformation in which economic activity was isolated and the distinct economic motivation of the feudalist system was usurped. This transformation allowed economic activities to be controlled, regulated and directed by market forces, whereby supply and demand were coordinated by the price mechanism. In this sense, a market for land was developing (Plata, 2001).

THE RECENT BRAZILIAN EXPERIENCE

At the end of 1994, 30 years had passed since the promulgation of the Land Statute. Representing a period of more than a generation, this should have been more than enough time for social transformation to have taken hold, given the experiences of other countries. However, the Brazilian land reform issue had still not been adequately addressed. The results of the agrarian reform programme up until 1994 had done little to ameliorate the miserable living conditions of millions of Brazilians who had little or no land of their own. Up to 1994, around 300 000 families had benefited from federal aid and state land agency agrarian reform and settlement projects. However, an estimated 4 million families did not benefit from these programmes.^[2]

TABLE 1
Goals reached by the government in the agrarian reform project (1995 - 2001)

Source: INCRA (2002). Balanço da Reforma Agrária e da Agricultura Familiar.

The agrarian situation after 1994 deteriorated as a result of land occupation and land seizure by the Instituto Nacional de Colonização e Reforma Agrária (INCRA), both of which resulted in increased conflict, which culminated in the slaughter of 19 rural workers in "Eldorado de Carajás". This grave situation provoked national outrage and forced the federal government to accelerate the processes of land distribution as well as directly address the plight of landless agricultural workers. After 1994, a favourable political climate facilitated modification of the land market, which increased its efficiency. This was in large part a result of the effective commitment of the Executive to institute policy change in response to the strong social pressure created by the "Marcha dos Sem Terra" (Landless March) in April 1997.

Land conflict and the lack of more aggressive land policies encouraged important social movements to employ land occupation as a third way.^[3] This process, led by the "Movimento dos Sem-Terras" (MST; Movement of the Landless) gained social support in its initial phase. This forced the federal government to renounce its previous policy and institute a land redistribution process never before seen.

Between 1995 and 2001, the federal government distributed 18 737 000 ha of land, which benefited 584 301 families. The cost per family decreased from $R19 412 to $R8 294 from 1995 to 2001, and the average price per hectare decreased from $R382 to $R264 during the same period (Table 1). After 2000, the process suffered a slight reversal.

The data in Table 2 are based on a sample from agricultural studies that evaluated INCRA land distribution policies in the period between 1997 and May 1999. The data seem to indicate the following results from the agrarian reform measures instituted by Fernando Henrique Cardoso's government: of total land distributed, only 21.1 percent had been in use, 54.0 percent was non-utilized land, another 24.9 percent of land was set aside for permanent preservation, legal reserves or was inappropriate for farming, meaning that only 75 percent of the land initially allotted could be used for purposes of land reform. These unused lands were either of very low quality or would have required a prohibitively high investment to make them productive. Distribution of this land would have been untenable either from the standpoint of the beneficiaries or the government as it would have required a huge investment in rural credit to do so. On average, the lands that might present simple conservation problems but that could still be cultivated (Types I and II) represented 13.66 percent of the total; land with conservation problems of medium complexity (Type III) 41.56 percent, and lands with complex conservation problems but still cultivatable (Type IV) 22.38 percent of the total. Lands unsuitable for intensive use, but still adaptable for pastures and/or deforestation and/or wildlife (Type V), represent 7.28 percent of the land. Therefore, a great number of the beneficiaries who receive these lands will require significant investment in technology and credit to make these lands productive. The average agronomic note attached to distributed lands (0.56) confirms these findings.

TABLE 2
Brazil - Average indexes in the process of land evaluation for land distributed as a function of agrarian reform (implemented by INCRA)

Source: Ficha Agronômica de Desapropriação, INCRA. July 1999.

The price per hectare of land designated for distribution, $R21 500, is considered low in comparison with the prices shown in Table 1. However, taking into account that the average type of land distributed is Type V, this price is not as low as one might expect.

The average expense per family in the distributed areas, $R9 782.00, indicates only the value of the land parcel, whereas the state must account for the entire economic value embodied in both the land and those squatting on it.

Finally, these data bring into focus the severity of the problems that both the beneficiaries of the agrarian reform and the state will need to confront. Property rights are just one constraint on agricultural productivity. The state must ensure through its policies that the squatters have access to other markets, including credit, products, inputs and technology markets. Land distribution is only the first step in the process of increasing opportunity for and improving the lives of the rural poor.

Regional impacts of the settlements

A recent study promoted by the Núcleo de Estudos Agrários e de Desenvolvimento Rural (NEAD) (Heredia et al., 2001) brings to light valuable information on the regional impact caused by agrarian reform and its resettlement policies in Brazil.^[4] The strong concentration of settlements in these regions has its origin in the crisis of the local agrarian systems, characterized by extreme poverty, a rising number of endemic social conflicts, and the emergence of social movements and social organizations. From the statistical sample, in 95 percent of the cases, extrajudicial land dispute resolution was the basis for current claims. Notable examples of this phenomenon can be found in the northeast sugar cane region where there is a crisis in the sugar cane system, in the cacao zone in the south of Bahia where there is a crisis in the latifundia cacao farm system and in the Ceará hinterland where there has been a drawn-out crisis in the cotton industry in addition to droughts and a farming crisis. Data on the origin and locale of squatters' previous residences corroborate these findings: 68 percent of the sample group previously inhabited agricultural areas, 70 percent inhabited the same city or cities in the same region as before and 68 percent were born in the same city or cities in the same region in which they remained. Added to the fact that about 15 percent of the sampled settled population was born after the settlement agreements, we can conclude that local populations with historic attachment to their land is in the range of about 84 percent.

Although variable with respect to region, the demographic impact of the settlements cannot be disputed. In the region of Distrito Federal (DF), for example, the population displacement caused by the settlements seems to be diluted by the greater regional migratory trends into the capital and its surrounding areas.^[5] However, "the population in the rural settlements is equivalent to 23.73% of the total rural population (...)". This proportion reaches 80 percent in Natailância, MG, 65 percent in São João d'Aliança, GO, and 63 percent in Riachinho, MG. In Riachinho the population of the settlements is equivalent to 68 percent of the urban population and in Flores de Goiás it is greater than the urban population, which represents 49 percent (Heredia et al., 2001, pp. 225 - 226).

In some regions, the settlements have absorbed part of the marginalized urban population: in the region of DF, 34 percent of the settlements' population previously lives in the urban area; 22 percent in the Southeast of Pará, and 28 percent in the South of Bahia. In some cities, as in the west of Santa Catarina, the settlements resulted, in the latter half of the 1990s, in a migration from urban to rural areas - a reversal of the earlier trend. The rural population of Abelardo Luz, which decreased between 1990 and 1996, started to increase by approximately 6 percent annually between 1996 and 2000.^[6] In Passos Maia, in the same region, after a reduction of more than 21 percent of the total population and 25 percent of the agricultural population during the 1980s, the trend was reversed, in large part as a result of the settlements. The agricultural population experienced an increase of 30 percent and the total population an increase of approximately 35 percent.^[7]

Social pressures

The consequences of this phenomenon are important. Previously, the vast majority of public investment in infrastructure (roads, electrification, water supply, etc.) was directed towards urban areas and their enormous populations. Indeed, as most of the settlements had their origins in conflicts, it was to be expected that initially part of the local society considered inhabitants of the settlements as outsiders or potential intruders.

However, in many regions, the settlements had gradually produced population conglomerates^[8] that attracted the attention of local public agencies, and in some cases the majority of their investment. Thus, as the authors of the study point out, "in many cases, the creation of the settlements results in the amplification of the infrastructure demands and pressure on the city halls, responsible for the installment of several, as well as on the state government." (Heredia et al., 2001, p. 218).

Some data highlight this phenomenon. In 86 percent of the settlements there is a school, most of which (75 percent) were created after the settlements and from claims of the squatters (71 percent).^[9] More than 90 percent of the children between 7 and 14 years go to school, 63 percent of the young between 15 and 19 years, and 19 percent of those between 20 and 29 years. This compared with the previous situation in which 32 percent of those over 30 years of age have never regularly attended school.

In addition, in 64 percent of the settlements there are educational projects for young adults in progress, in particular projects of the Programa Nacional de Educação na Reforma Agrária (PRONERA; National Programme of Education in Agrarian Reform), sponsored by the Rural Development Ministry and nongovernmental organizations (NGOs). More than 20 percent of the squatters had taken technical courses after the establishment of the settlements (fewer than 4 percent had taken them before). As in most rural areas of Brazil, few settlements have health care systems, and even where this is the case, the daily presence of health professionals is rare (only four settlements). Therefore, the squatters seek medical attention in urban areas, increasing the pressure on urban services. However, in 78 percent of the areas covered by the project there are health agents who are subsidized in large part by local governments.

For the majority of families, the settlement also represents the first chance for access to credit, the banking system and the financial market: previously, 93 percent of the population interviewed had no access to the credit system; in 1998/99, 66 percent had some form of credit.^[10]

Access to the settlements remains poor,^[11] because a considerable part of the investment in infrastructure still remains to be made. However, in 66 percent of the cases there is collective transport inside the settlement at least once a week (and in 42 percent of the cases this is many times a day). One particular case is of several cities on the northeast of Zona da Mata, where old roads have been abandoned and new ones constructed by the local government to facilitate access to the settlements.

Living conditions in the settlements have also improved significantly: 74 percent of the squatters live in masonry residences (against only 39 percent previously), and 78 percent of the settlements have electricity (in 53 percent of the cases, in most or all of the plots).

For 59 percent of the families, the communal spaces are important meeting places, followed by residences (53 percent) and religious areas (only 18 percent of the cases). The settlements also provided the impetus for new organizations (associations, cooperatives, nuclei, etc.) and have contributed to strengthening organizations and social movements involved in political mediation. In all the studied regions, there are cities in which the squatters participate in the city councils responsible for rural development, agriculture, etc. In many cases, squatters have stood for public office (council members, member of the House of Representatives) and, in some cases, had even been elected mayor.

This inclusion presents two important achievements in terms of citizenship and human development. On the one hand, this population, marginalized before establishment of the settlements, began to have access to public politics and gained social and political recognition. On the other hand, the settlements and their organizations have gradually been taken into account by the cities, business interests and local political forces and they have started to exercise an active role in the definition of politics and public investments. In many cases, the local government has redefined its priorities. In the city of Abelardo Luz, in Santa Catarina, "municipal controllers affirm that, today, 41% of health and social assistance and 55% of funds pertaining to school transportation have been targeted at squatters." (Heredia et al., 2001, p. 218).

The remodelling of the agrarian system and the dynamic process of economic life

The impact of the settlements in the concentration of land property is variable, according to the specificities of each region. Available tax data do not yet permit us to measure the observed changes with accuracy.^[12] However, the study brings to light some interesting comparisons. In the studied regions, the settlements have accumulated less than 12 percent of the total area of farming establishments. This percentage is smaller in the South of Bahia and in the area of Distrito Federal (3 and 5 percent, respectively), but reaches 23 percent in the hinterland of Ceará and 40 percent in the Southeast of Pará. This gives some perspective of the incorporated territory in terms of the agrarian reforms in those regions.

TABLE 3
Proportion of the area of a settlement in the total area of the regions

Dismemberment of large properties (more than 500 ha, in 76 percent of the cases) also resulted in extension of and greater weight given to family farms in the local rural system. In some regions, the area occupied by settlements, with their characteristic family-based agriculture, represents up to 100 percent of the total area examined, as Table 3 demonstrates. In extreme cases, as in some cities of the sugar cane zone of the northeast, the area comprising establishments was increased by a factor of 4 (Pedras de Fogo, Cruz Espirito Santo), 5 (Agua Preta) or 6 (Maragogi).

By contrast, the study also found evidence that only a small percentage of plots change hands: in 91 percent of cases, the families who originally established the plot are still responsible; and in 96 percent they alone manage their own plot.

One of the first consequences of the transformation from latifundia farms into family-owned plots is the productive reorientation of the local rural system. In some regions, cultivation predominated in the latifundia farms. In the settlements, by contrast, there is greater variation in terms of production.

Aggregate data of areas affected by land reform confirm this finding. Six groups of products characteristic of family agriculture represent more than 70 percent of the gross of value product (GVP), excluding livestock and meat: milk and derivatives; cassava and flour; maize; beans; eggs; and rice. Some products of regional importance represent about 20 percent of the GVP: pineapple, soy, sugar cane, tobacco, passion-fruit, potato, pumpkin, sweet potato and cotton. More than 70 other products comprise the remaining production of the settlements.

Farm products are also diversified. Dairy cows are present in 52 percent of the settlements; poultry in 80 percent; swine in 34 percent; and goat and sheep in 21 percent (in 74 percent of Ceará State settlements). One IBGE research project in the studied area in 1999 found that dairy cows represent 8 percent, swine 9 percent, poultry 14 percent, and goat and sheep 25 percent.

Many of these products are destined for agro-industrial processing. A part of this processing is carried through by agroindustries (such as cotton, in Ceará, or milk, in almost all of the regions), but a relevant part is performed in the settlement or in its closest neighbouring regions (cassava, cheese and flour, in the area of the Distrito Federal; cassava, honey and flour, in the South of Bahia; cassava and flour in the Southeast of Pará and in the northeast sugar cane zone). This has immediate consequences for the economic activities related to rural production, inside and outside the settlements.

The importance of having a free market and increased supply for local consumers is clear. In the southeastern section of Pará, this transformation to a market system has resulted in greater food security. Consumers interviewed in the free markets of Pedras de Fogo, Paraíba, affirmed that the settlements had resulted in increased supply (and choice) of, and lower prices for, food. Thus, diversified production translates into improvements in food safety and supply and increased choice not only for the squatters but also for the urban population.

Evidently, the use of primary input is not always a good indicator of the sustainability of production systems,^[13] but it can be an indicator of the secondary markets created by the settlements in terms of industry and services. Only 18 percent of people interviewed said they had not used any type of agricultural input. In approximately 53 percent of the plots, seeds supplied externally were used; in 42 percent, veterinary help; in 40 percent, insecticides/ pesticides; in 37 percent, farm chemicals; and in 18 percent, organic fertilizers.

Squatters are increasing their stake also in the financial markets. Financing provided to them in 1998/99 represented 12.5 percent of the total provided to agriculture in the studied cities. This percentage is lower in the South of Bahia (4 percent) and in the zone of Distrito Federal (6 percent), but reaches 58 percent in the Southeast of Pará and 81 percent in the northeastern Zona da Mata.

The creation of jobs and improved income

Before establishement of the settlements, 30 percent of squatters had no access to land: 5 percent were unemployed, 11 percent temporary employees and 14 percent permanent employees. Fifty percent had only precarious access to land: 18 percent were partners or leaseholders, 17 percent were non-remunerated members of the family, 3 percent were small proprietors, and 12 percent had both precarious accesses to land and/or any type of job (permanent or temporary).

This situation changed radically with the emergence of the settlements: 84 percent of those interviewed confirm that work conditions have improved. In the 1 568 plots studied there are more than 4 765 people over 14 years of age who work, translating into an average of three people working per plot. If we consider all ages, there is an average of 3.57 workers per plot. Of these, almost 80 percent work exclusively in the plots (average of 2.6 jobs per plot). Only 1 percent of them work exclusively outside their plot. More than half of the squatters working outside their plot work in the settlement, and a significant part of them are in non-agricultural activities (50 percent in the South of Bahia and about 20 percent in Santa Catarina and Zona da Mata in Paraíba). Only 25 percent of those who work outside the settlement have only one job.^[14] Moreover, the squatters have created jobs for non-family members: in 36 percent of the plots people from outside the settlements were hired.

We are interested in jobs created directly by the settlement and inside the settlement.^[15] In less than 12 percent of the plots was there a loss of members as a result of work-related problems. By contrast, in 23 percent of the plots there was incorporation of new members of the family (first-degree relatives of the family head), resulting in an average of 2.4 relatives per plot (Heredia et al., 2001, pp. 235 and 236).^[16] We need to add to these direct jobs the non-agricultural jobs created or stabilized by the settlements (industries resulting from and for the settlements, public infrastructure implantation, services, commerce). Although, it was not part of the research objectives to measure with precision farming income and the total income of the families, we were able to assess the settlements' capacity to generate income. The following data were considered: "income from the work realized outside the plot; income from farming products, commercialization and other incomes or financial aid received." (Heredia et al., 2001, p. 426).

TABLE 4
Average family annual income and gross income 1999/2000 (*)

Source: Heredia et al. (2001, p. 435).

*Average income, considering the number of interviewed, including those without any income. To facilitate ease of analysis we eliminated fractions of a dollar. Family gross average income, according to the sample, is $R31 200 per month, slightly higher than two minimum wages, varying from $R11 674 in Ceará to $R43 872 in the southeast of Pará. Although there are strong regional variations, most of the studied population is above the poverty level, as demonstrated in Table 5 (the data concerning external jobs may be underestimated).

It is important to mention that the income from the commercialization considered here is an estimate of the potential monetary income of the plot and that the products for exclusive consumption have not been considered.^[17]

It is important also to note that, in particular in Ceará, but also in part of the northeastern Zona da Mata, agriculture had been affected by a long period of drought. Table 4 demonstrates that almost 70 percent of the gross income of the families comes from the plot. Retirement income is the second largest source of income for settled families (17 percent), something reported in previous studies of family farming and social welfare provision in rural areas.

Plot capitalization, as an indirect indicator of the income level of farming families, confirms these findings. Indeed, the capitalization occurs either through credit or when family income exceeds bare subsistence levels. The majority (67 percent) of productive installations had been constructed with squatter resources and 55 percent of the machines and individual equipment had been financed by them.

The settlements as a factor in development

The settlements have diversified their productive system and distribution networks, increasing food supply and the consumption of agricultural and nonagricultural goods and services, as well as of durable goods.^[18] This process has created a considerable number of direct and indirect jobs, both agricultural and non-agricultural, which has increased the demand for and the investments in infrastructure and basic services (health, education, transportation), resulting in the diversification and transformation of the economic life of the cities.

Created during a crisis within the local agrarian systems, the settlements diversified and transformed the local economy and extended the chances of economic and social inclusion, not only for the settled families. The settlements became a development factor.

NEGOTIATED AGRARIAN REFORM: RECENT EXPERIENCES

Recently, some agrarian credit programmes have aimed to facilitate land access for those without land or with little land. Their objective is to provide credit for the small buyers, who normally would not have access to financial markets. The World Bank has been stimulating this kind of agrarian reform programme, and intends to grant long-term credit (for acquisition of the land) and partial subsidies (for investments in infrastructure and productive projects) to low-income beneficiaries.

According to some researchers, the supporting reasons of these programmes are:

• flow of information between buyers and sellers,

• competition between sellers,

• competition between potential buyers in the financing of productive projects,

• voluntary negotiation of the land price among groups of sellers and buyers,

• technical assistance to poor workers in the design of projects,

• subsidies to obtain partial financing of projects, including land purchase, and

• monitoring system.

These programmes had been designed to complement and eliminate some gaps in the agrarian reform process. The advantages of these programmes are explicit:

• in the higher freedom of the beneficiaries, because they can choose the land they prefer and negotiate its price;

• in the suppression of the state agency intervention in land selection and negotiation - surpassing the bureaucracy; and

• in the reduction of administrative costs and possibility of transference of functions from the governmental agencies to the private sector, especially in the areas of project preparation and technical assistance to beneficiaries.

But they also include some risks, which might increase costs, such as:

• inequality in negotiation influence,

• non-flexibility in land offers, and

• difficulties recovering credit.^[19]

In Brazil, the programme began in Ceará, in 1996, through a project of the state government entitled Programme to Combat Agricultural Poverty, financed by the World Bank (Project São José - the Solidarity agrarian reform). In 1997, the federal government adopted the idea and initiated the implementation of a new similar project in five states: Bahia, Pernambuco, Ceará, Maranhão and Minas Gerais. Federal government resources^[20] assure the beneficiary associations of the recovery of long-term financing (20 years) for land purchase. After a grace period of 3 years for the first reimbursement, the loan has negative real interest (6 percent per year, reduced to 3 percent for timely repayment).^[21] A Loan Agreement with the World Bank assures the non-reimbursable financing of projects, productive investments or basic infrastructure, so that the communities can design the property and begin its development. Resources for the installation of families ("initial allowance for expenses") and the technical expertise contracted by the community are examples of these projects.^[22]

The project, decentralized and the responsibility of the states, allowed the settlement of more than 14 000 families, who acquired 370 000 ha. At the end of the project, in 2002, the goal of 15 000 settled families has been exceeded. To improve on this experience, the Ministry of Agrarian Reform created in 2001 the Credito Fundiario (Project of Agrarian Credit and Combat to the Agricultural Poverty), with the participation of the Confederação Nacional dos Trabalhadores na Agricultura (CONTAG; National Confederation of Workers in Agriculture), the goal of which is, in the first phase of the project, to settle 50 000 families in 3 years.^[23] The area covered by this project was extended to allow the incorporation of all the northeast states, Espirito Santo (southeastern region) and, experimentally, three states in the south (Rio Grande do Sul, Santa Catarina and Paraná).

In 1998, the federal government also created the Banco da Terra. Although its execution is decentralized and its actions are based on the principle of financing property acquisition, Banco da Terra has significant differences with the Project Cédula da Terra. One of the main differences is that the Banco da Terra does not have the express purpose of combating poverty, because it foresees the reimbursable financing of all its investments, including infrastructure, and does not limit beneficiaries to the poor. Whereas the projects Cédula da Terra and Crédito Agrário had fixed the maximum financing and subsidies to $R15 000, Banco da Terra allows financings of up to $R40 000. Some of the resources of the Banco da Terra had been applied in the three states of the south and the average values of the transactions exceed considerably those verified in the Project Cédula da Terra.

However, these three land acquisition financing programmes allow lands that cannot be dispossessed in the traditional way to be incorporated into the process of agrarian reorganization (properties of less than 15 fiscal modules or productive properties of more than 15 modules). This represents about 40 million hectares in Brazil.

Preliminary results of the Project Cédula da Terra

It is still too early to give a thorough evaluation of the results of Cédula da Terra. A preliminary evaluation of its impacts was carried through by a team of researchers, led by A.M. Buainain between 1999 and 2000, which allowed them to determine the profile of the beneficiaries and to evaluate some aspects related to the strategy of action of the project.

The impacts relating to improvement in the quality of life are comparable with those verified in the settlements of the INCRA. In Pernambuco, for example, 89 percent of the beneficiary families live in masonry houses, versus 78 percent before the project (Vital et al., 2001). The consumption of durable goods was small ($R140 per family on average), but reasonable, if we take into account the short period since the beginning of the project versus their initial situation.

TABLE 5
Average of annual gross family income according to the range of minimum wage 1999/2000

Source: Heredia et al. (2001, p. 435).

The minimum wage was, for the period of the survey, $R15 100.

TABLE 6
Prices of farmland from FGV, hectare cost on Cédula da Terra and dispossession cost from INCRA

Source: Reydon and Plata, 1998a.

(a) Real price of farmland, Boletim Estatist. do Centro de Estudos Agrícolas IBRE/FGV (June 1998 = 100).

(b) Average cost per hectare, according to the database of Cédula da Terra, February 2000, NEAD.

(c) Average price of the land dispossessed by INCRA, per hectare, 1996 - 1998, Departamento de Finanças - INCRA. In Gasques, J. and Conceição, J.

Demanda de Terra para a Reforma Agrária no Brasil Box 5, p. 38, Brasília. November 1998.

Production was affected by a severe drought in the area (including the state of Minas Gerais, where the project was implemented). Moreover, at the start of the project (until the middle of 2000), there was a delay between the acquisition of lands and the release of resources for the community investments, either by the project or the Programa Nacional de Fortalecimento da Agricultura Familiar (PRONAF).

Even so, the monthly average income of the families interviewed in Pernambuco increased from $R11 300 to $R20 600 in 1999/2000 (Table 5). Families' gross income increased from $R1 361 to $R2 474 per year, an increase of more than 80 percent. A significant part of this income is still derived from activities outside the plots, as was the case before settlement, because 35 percent of those interviewed still worked outside the settlement. On average, income acquired from external activities represents 28 percent of the total family income, whereas income derived directly from farming activities represents 15 percent of the total.

The impact of the project in the land market

Table 6 indicates the costs per hectare of land redistribution under agrarian reform, the cost per hectare of the Project Cédula da Terra (PCT) and the value of the farmland estimated by the Fundação Getúlio Vargas (FGV). We conclude that the Project Cédula da Terra did not cause overvaluation of the lands, nor established prices to be greater than market standards (inferior indexes to those used by the FGV) or those established by INCRA. On average, these are 62 percent less in Maranhão, 66 percent in Ceará, 14 percent in Pernambuco, 43 percent in Bahia and 49 percent in Minas Gerais.^[24] These data seem to confirm some of the hypotheses that justify projects such as the Cédula da Terra. These programmes, when restoring direct negotiation and single payment, with no resource to justice:

• allow the reduction of transactions and, therefore, the sum of resources owned by the proprietors;

• make it possible to understand movements in the price (falling) of the land in the short term; and

• reduce the time between identification of the property and the settlement of the families.

By contrast, these projects do not depend on the capacity of direct state action in all the phases of the process. This allows these programmes to speed up the productive development of the settlements, reduce the cost of the investments in infrastructure and accommodate higher numbers of beneficiaries.

THE AGRARIAN REFORM AND SOCIAL CAPITAL FORMATION

Social capital is a basic condition for the viability of economic and social development of marginalized populations. This has become a commonly accepted principle through the endorsement of internationally well-known researchers. We aim to add to this an understanding of the effectiveness of social capital for the process of social development in the context of the settlements resulting from agrarian reform. We use the definition of social capital as used by Bourdieu (1998: 65):

"Social capital is the set of current or potential resources related to the ownership of a durable net of relations more or less institutionalized of cognition and inter-recognition or, in other terms, the entailing to a group, as a set of agents who, are not only endowed with common properties (that can be perceived by an observer, other than themselves), but are also joined by permanent and useful linking. These relations are irreducible to objective relations of proximity in the physical space (geographic) or in the economic and social environment because they are established in exchanges both material and symbolic which establish and perpetually assume the recognition of this proximity."

In this definition, Bourdieu strengthens two basic fundamental approaches to our investigation: first, it isolates the group identity, as base of the formation, appropriation and durable ownership of the social relations that form the essence of social capital; second, the appropriation concept works as the process "to concentrate in the hands of a single agent the totality of the social capital that establishes the existence of the group", establishing the dialectic contradiction in the interior of the organizational process.

Contradiction between coordination and base in the social organizations

Although we recognize that the more organized rural workers are, the larger their access to institutional resources, the better use they make of these resources and the more rational and effective is their capacity for production and social inclusion, it is indispensable to analyse some contradictory aspects of how social capital is organized between workers. For this we consider the importance of the autonomy of base organizations, related to the hierarchical structures of coordination on the state, regional and national levels, as the basis for the construction of democracy and the citizenship of the local participants.

In our analysis of the settlements of agrarian reform, the fragility of the social organization of squatters is evident where social capital could provide an important differential in economic and social development, as evidenced in a recent study on the factors of success and failure of the settlements (Bittencourt et al., 1998). The power and the capacity of organization of the rural workers at the state, regional and national levels do not correspond to the settlements. The broader the scope of the organization, the stronger its structure.

In that sense, autonomy as related to the base organizations cannot be surpassed by the maniqueistic speech that equates local singularities to collectivist equality and associates the universal right to individuality with the individualism and social isolation of the elites. We cannot accept the development of rural settlements without considering the construction of this autonomy and how it relates to the characteristics of each situation and to individual differences. At the same time, there is no way to increase the power resulting from worker coordination without the reinforcement of this base autonomy rooted in the democratic process. As seen in practice, the strengthening of local autonomy is fundamental both to communitarian development and to the reinforcement of workers' social organization in broader terms.

AGRARIAN REFORM TO AGRICULTURAL DEVELOPMENT: A NECESSARY EVOLUTION

In Brazil the vision that agrarian reform "annihilates" the old order (as though Brazilian capitalism were dominated by agrarian oligarchies, still prevalent in academic sectors, in part of the catholic church and in important social movements) has failed since its inception. Its doctrine would be to eliminate the latifundia farms and consequently to establish a new socialist order (not democratic). It is intellectually important for the progressive political parties to try and change this vision, even without success, because otherwise they suffer a type of "blackmail" from the most radical sectors, obliging them not to voice a new position, and relegating their argument to purely academic discussions.

Recent evidence shows a notable growth in non-agricultural occupations in Brazil, a phenomenon observed in the majority of Latin American economies. Rural Brazil comprises about 4 500 cities, with a population of 50 million inhabitants, with new and dynamic economic possibilities for both agricultural and non-agricultural pursuits. Together with the almost innate entrepreneurship of Brazilians, especially the poor, we have ahead the unique opportunity to create millions of productive occupations. The old notion of agrarian reform has to give way to one that, still respecting the importance of land redistribution, may instigate rural development in a broader territorial approach that not only fights poverty, but also builds citizenship.

The territorial focus of rural development

The problematic issue of territorial development demands that three main subjects are taken separately: resetting local territories, local productive systems and the environment. In practice, it is impossible to deal with each of these in isolation. However, it is in trying to separate them that we can explicitly understand what is meant by territorial development.

Many nations adopted their current territorial structures long ago and, as such, these structures do not address the real necessities of modern economic growth, or of sustainable development. In this context, urban/rural relations no longer correspond to the obsolete dichotomy between the city and rural areas. During the twentieth century, the development process simplified this dichotomy, substituting it for a variable geometry in which the agglomerations and the microregions began to be more and more crucial. These two categories can be combined in the metropolitan regions, but can have different relations in other areas. Microregions that involve a non-metropolitan agglomeration tend to be essentially urban. But microregions that only involve urban centres and/or "rurban" villages can be relatively rural, or even essentially rural, when less artificial systems predominate.

The expression "resetting of the territories" therefore means the necessity of new institutional forms of coordination, management or governing, of the agglomerations and the microregions. This is a challenge that in many countries has been called "intercommonality". In Brazil this corresponds to the promotion of intermunicipal microregional unions, as for example with the associations of Santa Catarina, or the diverse types of trusts existing in the country. When an agglomeration or a microregion achieves a high enough degree of cohesion and organization that it is capable of formulating and adopting a plan of local development, it is inevitable that it perceives the strategic importance of two decisive factors: the local productive system and the environment (Veiga, 2002).

A less aggregated approach to territorial configuration in Brazil reveals a trend that should not be ignored by policy-makers. If we add to the urban classification all of the intermediate cities those villages of "rurban" type that might become urban centres, we obtain a total of 1 022 cities, in which almost 118 million people lived in 2000. In this expanded subgroup, population growth between 1991 and 2000 was approximately 20 percent, perhaps stemming most notably in increases from non-metropolitan agglomerations and urban centres. However, this does not mean that in all the remaining cities - small and essentially agricultural - there has been a population leakage. This, in fact, has happened in half of these cities, whereas in a quarter there has been an increase in the population of 31.3 percent; this is greater than for urban Brazil, which grew by 18.8 million (19 percent) between 1991 and 2000 and more than double the population growth of Brazil as a whole, which was of 15.5 percent over the same period. The factors that attracted people to these 1 109 cities with rural characteristics and led to the 31.3 percent increase in population from 1991 to 2000 are largely unknown. It is not possible for us to consider more than 90 percent of Brazilian territory as a whole, 80 percent of its cities and 30 percent of its population as the rest comprise residue from the urban - industrial revolution of the second half of twentieth century. It is not possible to treat the country as if there were 4 500/5 000 imaginary cities.

However, what matters is not a comparison between the demographics of urban, agricultural or intermediate localities, but rather to understand that the future of these populations will vary according to their interconnectedness, and to identify the vocations of the territory they share, to formulate a plan of local development, and to make possible its financing with the essential support of the national government. It is clear that the microregions that do not form part of an agglomeration will be less prepared to face this challenge. We can conclude it is important to develop the federal programme tuned in particular to the promotion of connections between microregional cities with small populations, i.e. the programme that has especially focused on the sustainable development of rural Brazil (Veiga, 2002).

Strategic recommendations for land policy

Our experience with the reforms of the 1960s has shown the necessity of functioning public institutions, and that not enough has been done with this in mind.

In order to transform any society, the entire population must be involved. Institutions deriving their legitimacy from participation and becoming involved in the development process are necessary as legitimacy and participation will enable the development strategy to be adapted to a particular country and will ensure commitment and involvement in the long term, in turn increasing the prospect of sustainability (Stiglitz, 1998).

With regard to land policies, participation is even more important. Access to land for poor workers constitutes a unique opportunity for the social trajectory of these groups in Brazil. Some recommendations to improve the efficiency and the effectiveness of land policies in the country are as follows.

Update the legal system

Land redistribution and registry mechanisms still result in incorrect land pricing. This constitutes an abusive transference of wealth from landowners and intermediaries and results in difficulties for the beneficiaries in terms of repayment for lands.

Flexibility in obtaining and transferring land

This means regulating: both land redistribution and the recovery of illegally taken public lands (according to pre agreed requirements); the use of direct financing for purchase by interested parties, including incentives for lower prices offered in the negotiation process; leases with minimum limits on repayment deadlines and incentives for longer-term contracts (for example, the exemption of the territorial income tax and/or territorial for payback dates greater than ten years, with privilege rights); fast and simplified administrative titling for ownerships of more than five years and a fixed limited area of familyowned property.

Local governments

There are important omissions regarding territorial organization in legislative systems. Despite restrictions for environmental concerns, there are no mechanisms that would give rural workers the incentive to value their environment.

Modernization of registration systems and taxation

Deficiencies in the rural registration systems have led to difficulties in the collection of taxes, imprecision in public registry, and fraud and illegal seizure of lands. Accurate registry mapping has been prohibitively expensive. Therefore, only rich countries have been able to develop these systems; generating information that allows the creation of necessary legal security for the development of the land market is necessary. The availability of registry information in the cartographic plan would allow infrastructure planning that could support regional development.

Currently, as a result of technological advances and the use of satellite data, it is possible to make measurements and identify landmarks at much lower costs than previously. It is possible, for example, to make a topographical perimeter survey, utilizing GPS data in a fast and cheap manner. This digital information can be transmitted to a small central collection point, enabling the construction of a digital database that with complementary field and office surveys will allow the creation of a multipurpose Space Information System.

Modernization of the rural registration cadastres, together with agricultural regularization programmes, will allow consolidation of information and provide conditions necessary for a better functioning land market.

For land taxation, if the cadastral modernization is adopted, Venezuelan law should be taken as a guideline towards a system based on the potential use of the land (Venezuela has been investing in a digitalized system for many years).

This will work as a parallel incentive for better use of land. Decentralization to an exclusive municipal government is another recommendation based on historical experience and arguments that proximity and interest will result in better administration

Instruments adapted to circumstance^[25]

Land access can be an effective strategy to alleviate poverty; however, its success requires particular conditions.

Land access does not ensure release from poverty (it is necessary but insufficient alone). Other conditions are necessary, such as market competition, remuneration higher than the cost of labour, and cheaper food than could be bought on the open market.

The requirements of land, infrastructure and access to proper financing are factors that will make the process of land redistribution viable, the objective being to extend family-unit agricultural production.

Poverty alleviation and subsistence farming

The idea that rural workers would provide for their subsistence only through farming activities led to the creation of land policies sensu stricto. The reality is that farm and off-farm activities depend on the availability of assets and the context (work opportunities inside and outside the property). In this sense, land redistribution in subfamiliar units must be a part of the strategy for rural development programmes and projects, and in particular for land access policies.

Multiple forms of land access

Inheritance or legacy, purchase and sell, illegal taking of public or private lands, lease and partnership, agrarian reform, fruition, etc. can each be improved and made more effective, with specific policies and programmes; agrarian reform is one means to achieve this.

Multiple types of agrarian reform

The different types of agrarian reform should be understood as complementary.

Four conditions of success for the implementation of a strategy of land access for the rural poor are on the supply side: political factualism (an excellent measurement parameter), and fiscal viability (the lowest costs and subsidies to obtain a broad coverage of the social demand); on the demand side: incorporation of those who are poor (who have access to a minimum of assets necessary to participate in the programme), and competitiveness of beneficiaries (economic sustainability with their insertion into the market).

Expropriations based on the social function of the property

This requires workable policies and strong social movements. It is important where land distribution is highly unequal, there is unused land and there are many people who own no land. Government action is essential to prevent violence and reduce the costs of dispossession. As expropriation is still the main instrument of land redistribution practised in Brazil, its use can be improved by the introduction of mechanisms that control the value of the financial indemnities.

The combination of public payment in money and public shares can introduce a bargaining element for administrative agreements. Judicial arbitration should be used only in exceptional cases. This obviously demands transparency and ample social control.

Redistribution through land market mechanisms

Innovative aspects: community participation; incentives for the promotion of bargaining; self-election processes of the beneficiaries; smaller groups and greater availability of social capital; lack of conflict; better insertion in the local society. Risks: local power, increase in the land price. The Brazilian experience already shows the risk reduction that arises from greater transparency and social control. Restricting use of this instrument only to nondispossessed land can be a limiting factor in the near future given some situations, and defined regions may be the best solution.^[26]

The leasing market has been known as an option for increasing land access in Latin America and Brazil. Lack of regulation and control over contracts explains why this avenue has not been pursued more often. This can be observed mainly in countries with evident inequalities in wealth distribution (land) and income. Even then, the historical experience of (now) developed countries suggests that this instrument can be a step towards land access (property), if the conditions identified in this document are met.

What to do

To complete the initiated agenda

The economic and social costs of agrarian reforms are high. Time negatively affects those beneficiaries who cannot be competitive if there is no regulated and accepted access to land (titling). In Brazil, despite some improvements, there is a "deficit" of infrastructure in the settlements that hampers development and the consolidation of productive activities. It is necessary to overcome the preconception that the workers, if they become the owners, will sell their lands. Actually, whatever the restrictions might be, there will always be a market (formal or informal). Its intensity should be the concern of public authorities, because they can demonstrate the lack of sustainable conditions (economic, environmental, etc.).

Land access and competitiveness

It is necessary to incorporate in agrarian reform interventions the "new" approaches of rural development: regional growth (context) and economic incorporation of the poor (active). Beneficiaries need to control the minimum levels of productive assets: experience; machinery, livestock; access to support institutions and markets.

Capitalize urban interests in environmental issues to negotiate land access

Water quality, combating atmospheric pollution, biodiversity and recreation are subjects related to environmental services that may affect land payment, depending on the Rural Development Plan in a particular locality. In the case of poor environmental use, possibile expropriation should be examined.

BIBLIOGRAPHY

Banerjee, A., Getler, P. & Ghatak, M. 1998. Empowerment and efficiency: economic analysis of tenancy reform program in India. Brasília. Paper presented in Seminário Internacional: Distribuição de Riqueza, Pobreza e Crescimento Econômico. NEAD-World Bank. 14 - 17July 1998. 42 pp. (mimeo)

Bardhan, P., Bowles, S. & Gintis, H. 1998. Wealth inequality, wealth constraints and economic performance. Brasília. Paper presented in Seminário Internacional: Distribuição de Riqueza, Pobreza e Crescimento Econômico. NEAD-World Bank. 14 - 17 July 1998. 72 pp. (mimeo)

Barro, R. & Sala-I-Martin, X. 1995. Economic growth. New York, USA, McGraw Hill. 539 pp.

Barus-Michel, J. 1987. Le sujet social. Étude de psychologie sociale clinique. Paris, Dunod. 209 pp.

Besley, T. & Burguess, R. 1998. Land reform, poverty reduction and growth: evidence from India. Paper presented in Seminário Internacional: Distribuição de Riqueza, Pobreza e Crescimento Econômico. NEAD-World Bank. 14 - 17 July 1998. 19 pp. (mimeo)

Bianchini, W. et al. 1998. Principais Fatores que Afetam o Desenvolvimento dos Assentamentos da Reforma Agrária no Brasil. Brasília, Convênio FAO/INCRA. (mimeo)

Binswanger, H. & Deininger, K. 1997. Explaining agricultural and agrarian policies in developing countries. J. Econ. Literature, 35(4): 1958 - 2006.

Binswanger, H., Deininger, K. & Feder, G. 1993. Power, distortions and reform in agricultural land markets. In J. Behrman & N. Srinivasan, eds. Handbook of development economics, Vol III. Amsterdam, The Netherlands, North Holland.

Bittencourt et al., 1998. Principais Fatores que Afetam o Desenvolvimento dos Assentamentos da Reforma Agrária no Brasil. Brasília, FAO/INCRA. (mimeo)

Bourdieu, P. 1998. The economy of symbolic goods. In Practical Reason. Stanford, USA, Stanford University Press.

Bruno, R. & Medeiros, L. 1998. Desistência e Abandono nos Assentamentos de Reforma Agrária. Brasília, Convênio FAO/INCRA. (mimeo)

Buainain, A.M., da Silveira, J.M. & Teofilo, E. 1998. Reforma Agrária Desenvolvimento e Participação: uma das discussões das transformações necessárias e possíveis. Brasília, NEAD.

Buainain, A. & Rezende, G.C. 1995. Structural adjustment and agricultural policies. In J.Weeks, ed. Structural adjustment and the agricultural sector in Latin America and the Caribbean. London, Macmillan Press.

Buainain, A. & Souza, H.M. 1998. Procera: impactos produtivos e capacidade de pagamento. Brasília, Convênio FAO/INCRA. (mimeo)

Carter, M. & Zimmerman, F. 1998. Can time and markets resolve agrarian question? Paper presented in Seminário Internacional: Distribuição de Riqueza, Pobreza e Crescimento Econômico. NEAD-World Bank. 14 - 17 July 1998. 41 pp. (mimeo)

Castoriadis, C. 1975. L'institution imaginaire de la société. Paris, Éditions du Seuil. 498 pp.

Chenery, H. 1974. Redistribution with growth. Oxford, UK, Oxford University Press.

Coleman, J. 1994. Foundations of social theory, 2nd edition. Cambridge, MA, USA, Harvard University Press. 992 pp.

de Janvry, A. 1985. Social structure and biased technical change in Argentine agriculture. In H. Binswanger & V. Ruttan, et al., eds. Induced innovation, pp. 297 - 323. Baltimore, MD, USA, The Johns Hopkins University Press.

de Janvry, A., Key, N. & Sadoulet, E. 1997. Agricultural and rural development policy in Latin America: new directions and new challenges. University of California at Berkeley, Working Paper no. 815. 43 pp. (mimeo)

Deaton, A. 1997. The analysis of household surveys. Baltimore, MD, USA, The Johns Hopkins University/World Bank.

Deininger, K. & Squire, L. 1998. New ways of looking at old issues: assets inequality and growth. Brasília. Paper presented in Seminário Internacional: Distribuição de Riqueza, Pobreza e Crescimento Econômico. NEAD-World Bank. 14 - 17 July 1998. (mimeo)

Eswaran, M. & Kotwal, A. 1985. A theory of contractual structure in agriculture. Am. Econ. Rev., 75(3): 352 - 367.

FAO. 1996. Mercado y Políticas de Tierras, by B. P. Reydon & P. Ramos. Rome.

FAO. 2001. Un Nuevo Trato para el Campo, by G. Gordillo. Rome.

Favereau, O. 1989. Valeur de Option et Flexibilité. In P. Cohendet & P. Llerena, eds. Flexibilité, information et décision. Paris, Economica.

Ferreira, A.H.B. 1995. O Debate sobre a Convergência de Renda per capita. Belo Horizonte, Revista Nova Economia, 5(2): 139 - 153.

Ferreira, P.C. & Ellery, R. 1996. Crescimento e Convergência de Renda entre os Estados Brasileiros. Rev. Econometria, 16(1): 83 - 103.

Fishlow, A. 1995. Inequality, poverty and growth: where do we stand? Proceedings of the World Bank Annual Conference Development Economics, pp. 57 - 60. Washington, DC, World Bank.

Grossman, S. & Hart, O. 1986. The costs and benefits of ownership. A theory of vertical and lateral integration. J. Political Econ., 94(August): 691 - 719.

Guanziroli, C. 1998. Land reform and globalization of the economy: the Brazilian case. Brasília, DF. (mimeo)

Hart, O. 1997. Firms, contracts and financial structure. Clarendon Lectures in Economics. Oxford, UK, Oxford University Press. 228 pp.

Heredia, B., Medeiros, L., Palmeira, M., Leite, S. & Cintrão, R. (coordinators). 2001. Impactos regionais de reforma agrária: um estudo a partir de áreas selecionadas. Rio de Janeiro, Brazil, CPDA-UFRRJ-Nuap/PPGAS/MN/UFRJ-Nead/ Iica.

Hoff, K. 1998. Wealth distribution, economic efficiency and incentives. Paper presented in Seminário Internacional: Distribuição de Riqueza, Pobreza e Crescimento Econômico. NEAD-World Bank. 14 - 17 July 1998. 31 pp. (mimeo)

Kageyama, A. 1993. A questão agrária: Interpretações Clássicas. Reforma Agrária, 23(3): 5 - 16.

Katz, C. 1990. Los Efectos de la Renta Agraria en Latinoamérica. Estudios Rurales Latinoamericanos, Colombia, 13(1/2): 49 - 84.

Kuznets, S. 1955. Economic growth and income inequality. Am. Econ. Rev., 45: 1 - 28.

Morin, E. 1995. Introdução ao Pensamento Complexo. trad. Dulce Matos, 2nd edition. Lisbon, Instituto Piaget. 177 pp.

Nelson, R. 1994. Coevolution of technologies and institutions. In R. England, ed. Evolutionary concepts in contemporary economics, 1st edition. Michigan, USA, Michigan University Press. 255 pp.

Pacheco, C.A. 1996. A Questão Regional Brasileira pós-80. Tese de Doutoramento apresentado ao IE-Unicamp. 334 pp. (mimeo)

Pagès, M. 1998. Le phénomène révolucionaire: une regression criatice. Paris, Descleé de Brouwer.

Plata, L.A. 2001. Mercado de Terras no Brasil: Gênese, Determinação de seus Preços e Políticas. Campinas University, São Paulo State. (Thesis).

Putnam, R. 1996. Comunidade e Democracia: a experiência da Itália Moderna. trad. Luiz Alberto Monjardim. Rio de Janeiro, Brazil, Ed. Fundação Getúlio Vargas.

Rangel, I. 1962. Desenvolvimento Econômico e o Empresário. Revista de Administração de Empresa no. 4.

Reydon, B.P. & Plata, L. 1998a. Evolução recente de preço da terra rural no Brasil e os impactos do Programa de Cédula da Terra. Brasília, NEAD. July.

Reydon, B.P. & Plata, L. 1998b. Políticas e Mercado de Tierras no Brasil. In Perspectivas sobre Mercados de Tierras Rurales en América Latina. Washington, DC, BID-Informe Técnico do Departamento de Desarollo Sostenible. 132 pp.

Romeiro, A. et al. 1994. O Mercado de Terras. Brasília, IPEA.

Sappington, D. 1991. Incentives in principal-agent relationships. J. Econ. Perspectives, 45 - 66.

Schiff, M. & Valdés, A. 1992. The plundering of agriculture in developing countries. Washington, DC, World Bank.

Shleifer, A. & Vishny, R. 1997. A survey on corporate governance. J. Finance, LII(2): 737 - 783.

da Silveira, J.M.F.J. & Kageyama, A. 1996. O Debate Regional na Atualidade. In Anais da II Reunião da Sociedade de Economia Política. São Paulo, Brazil.

da Silveira, J.M.F.J. & Rabelo, F.M. 1998. Estruturas de Governança Corporativa na Petroquímica Brasileira Pós-Privatização. Presented in XXVI ANPEC Meeting, Vitória, ES, Brazil, 9 - 11/12.

Stiglitz, J. 1998. Distribution, efficiency and voice: designing the second generation of reforms. International Conference on Wealth distribution, Poverty and Economy Growth. Brasília, World Bank/NEAD.

Teófilo, E. et al. 1998. Diretrizes para a Política de Desenvolvimento Agrário. Brasília, NEAD. (mimeo)

Turner. 1999. The formation of social capital. In P. Dasgupta & I. Serageldin, eds. Social capital: a multifaceted perspective. Washington, DC, World Bank.

Veiga, J.E. 1998. Pobreza Rural, Distribuição de Riqueza e Crescimento: a experiência brasileira. Paper presented in Seminário Internacional: Distribuição de Riqueza, Pobreza e Crescimento Econômico. NEAD-World Bank. 14 - 17 July 1998. 20 pp. (mimeo)

Veiga, J.E. 2002. Cidades Imaginárias. Estudos NEAD 6.

Vital et al., 2001. Estudo de avaliação de impactos socioeconômicos do programa Cédula da Terra em Pernambuco. Recife, Brazil, NEAD/Fadurpe/ IICA. (mimeo).

Williamson, O. 1996. The mechanisms of governance, 1st edn. New York, USA, Oxford University Press. 429 pp.

Zeller, M. et al. 1995. Rural financial policies for food security of the poor: methodologies for a multicountry research project. Washington, DC, Food Consumption and Nutrition Division, IFPRI. 53 pp. (mimeo)