Livestock Advisor, Agriculture and Natural Resource Department, World Bank

a Professor of Agricultural Economics, University of Manitoba, Canada

b Research Fellow, International Food Policy Research Institute, U. S. A.

Cattle production in the Brazilian Amazon has ignited one of the most visible and vocal environmental debates of the past decade, amid fears that the Amazon could swiftly disappear unless there was rapid change in attitudes and policies. The high rates of deforestation have been linked to environmental problems ranging from localized degradation of land through to global implications such as climate change. Although there are quite a few land uses in the Amazon, the predominance of pasture has drawn the ire of many environmentalists who say that cattle production is a bad deal; that the destruction of the largest remaining tropical rain forest to produce beef is not warranted by low productivity and questionable sustainability.

In the case of the Amazon, deforestation itself is often taken as prima facie evidence of poor land use, partly from the viewpoint that preservation of tropical forest is fundamental to the prevention of global climate change but also from a feeling that any agricultural use of the Amazon is doomed to eventual failure (Fearnside, 1997). A growing literature tries to identify the external and internal drivers of deforestation, with most attention focussed on the role played by external factors such as regional development subsidies and macroeconomic policies. However, less well recognized is that production systems themselves might be important factors in forest clearing. In this paper we try to emphasize both internal and external drivers. We do not accept the notion that deforestation itself is, prima facie, necessarily bad because of the role that agricultural production can play in economic growth and poverty alleviation, and the potential for production-system modifications that are environmentally-friendly.

Widespread poverty and economic stagnation over the past several decades, in Brazil and in neighbouring countries that share the Amazon, are pressing concerns that must be juxtaposed against the environmental consequences of Amazon exploitation (Psacharopoulos et al., 1995). The limited empirical evidence from colonization of the Brazilian Amazon suggests that human development indicators in the region have generally improved over time and that hard work by colonists is often, but not always, rewarded with increased wealth (Haller et al., 1996; Schneider, 1995). For many of the Amazon's rural colonists, the alternative is landless misery in rural areas elsewhere or deprivation in the slums on the margins of cities.

Therefore, we think it helpful to frame the main issues of Amazon development in terms of a “critical triangle” of goals: sustainability, growth and poverty alleviation (Vosti and Reardon, 1997). Trade-offs clearly exist, which underlie some of the policy choices taken or not taken in Amazon development. For example, poverty alleviation through Amazon migration was an explicit goal of the Brazilian Government during the 1970s (Schmink and Wood, 1992).

The objective of this paper is to stand back and highlight critical policy links for cattle production in the tropical forests of Western Amazonia. We take the perspective that the conversion of forest to pasture in the Brazilian Amazon is the result of a wide range of internal and external factors, and that indignation about subsidies and land speculation among large-scale ranchers has diverted too much attention away from other important issues of land use in the Amazon. Without diminishing the importance of the many environmental issues that have received widespread attention, we also suggest that the human consequences of Amazon development (economic growth and poverty alleviation) and their links to environmental sustainability have not enjoyed adequate consideration. Potential micro and macro policy levers, that might improve the performance of the livestock sector in the Amazon, have received scant attention outside the research activities of a few agricultural research bodies.

Extent of deforestation

Due to its sheer size and importance in terms of remaining primary forest a great deal of attention has been focussed on the amount and rate of deforestation in the Brazilian Amazon by the public media, scientists, environmentalists and government organizations. The best available deforestation data for the Brazilian Amazon come from the Brazilian Space Research Institute (INPE) remote sensing census.

In order to be clear about deforestation measurements it is necessary to establish a consistent base. Comparison across various studies is complicated because different authors use different definitions of the Amazon area and there is confusion about the amount of forest, savanna and water in those areas (Skole and Tucker, 1993). The Amazon drainage basin contains about 6 million km2 of rain forest, most in Brazil, and the remainder in the bordering countries along the Venezuelan/Guiana Highlands to the north and the Andes to the west (Colinvaux, 1989; Moran, 1993). Brazil defined the Legal Amazon as an economic and political planning region to administer economic development incentives in 1953, a region that covers the North Region states of Acre, Amapá, Amazonas, Pará, Rondônia, and Roraima, plus parts of Tocantins, Mato Grosso and Maranhão (Figure 1). The Legal Amazon contains a total area of about 5 million km2, of which about 4.2 million km2 was originally rain forest. INPE calculates deforestation on the basis of the Legal Amazon.

Figure 1: Brazil by State

Figure 1AC - Acre
AL - Alagoas
AP - Amapá
AM - Amazonas
BA - Bahra
CE - Ceará
DF - Distrito Federal (Brasîlia)
ES - Espírto Santo
GO - Goiás
MA - Maranhao
MT - Mato Grosso
MS - Mato Grosso do Sul
MG - Minas Gerais
PA - Pará
PB - Paraíba
PE - Pernambuco
PI - Piauí
RN - Rio Grande do Norte
RS - Rio Grande do Sul
RJ - Rio de Janeiro
RO - Rondônia
RR - Roraima
SC - Santa Catanna
SP - Sao Paulo
SE - Sergipe
TO - Tocantins

The Latest Deforestation News

Figure 2 shows the most recent estimates of deforestation for years in which INPE conducted a complete census of the Legal Amazon: 1978, 1988 through 1992 and 1994. Care should be taken in interpreting year-to-year changes in deforestation due to problems from cloud cover and the need to use images taken over an extended range of time.

Note also that a 1994 deforestation estimate for Amapá was not yet available at the time of writing so a value of 2,000 km2 was used as a crude 1994 estimate (the 1992 level was 1,736 km2). The data in Figure 2 include all land that has been identified as deforested through inspection of LANDSAT images, plus land in mature secondary forest known to have been deforested in earlier periods. The latest estimate is that roughly 471 thousand km2 or 11.2% of the Legal Amazon forest area had been cleared at least once by 1994. These data suggest annual deforestation rates in the 15,000–20,000 km2 range, a considerable rate of forest loss but well below the ad hoc guesses of 50,000–80,000 km2 (e.g., Myers, 1993; Repetto, 1990) that are frequently cited. Of that total at least 100 thousand km2, and perhaps more, had regrown to mature secondary forest.

INPE deforestation totals by state are shown in Figure 3 for 1978, 1988 and 1994. These graphs suggest that the annual rate of deforestation for most states was higher between 1978 and 1988 than between 1988 and 1994. In the latter period, official colonization programs, ranching subsidies and road construction were discontinued (or allowed to lie dormant). However, the rate and overall level of deforestation has remained very high in two states (Mato Grosso and Pará) that have substantial large-scale cattle sectors. More than half (53,000 km2) of the forest cleared between 1988 and 1994 occurred in these two states alone, with an additional 12,000 km2 of forest clearing in Rondônia.

Figure 2: Total deforested area - INPE Estimates

Figure 2

Source: INPE data on the World Web (

Figure 3: Latest estimates

Figure 3

Source: INPE data on the World Wide Web (

The “shifted cultivator” is frequently blamed for deforestation (e.g., Myers, 1993), but the significant role of large-scale cattle ranches in the Brazilian Amazon is the key difference between deforestation processes in Brazil relative to other areas. A wide range of farm sizes and production systems involving beef cattle, dairy cattle, annual food crops and perennial products can be found in the Brazilian Amazon (Mattos and Uhl, 1994; Villachica et al., 1990; Fearnside, 1983, 1989). Unfortunately, estimates of the amount of deforestation by farm size and type are hampered by a general lack of hard data and a careful evaluation of land use utilizing a statistically valid sample drawn from across the Amazon has not yet been done. Estimates range from a high extreme of 70 percent of deforestation being caused by large-scale ranchers to under 21 percent (Fearnside, 1993; Yokomizo, 1989; Reis and Margulis, 1991; Homma et al., 1995).

Cattle and amazon deforestation

Critics of Amazon development have highlighted government subsidies for large-scale cattle ranching and land speculation in colonization projects as particularly insidious causes of deforestation. Ultimately, most land that is cleared in the Brazilian Amazon ends up as pasture. Large-scale ranchers, of course, clear and plant vast tracts directly into pasture. But most land held by small-scale slash-and-burn farmers also ends up as pasture, first as a way to stretch out the productive life of land cleared for annual crops and later as a form of direct forest conversion. Largely overlooked as an explanation for the predominant place of pasture is the role played by rapid beef and dairy demand growth from occupation of the Amazon (Faminow, 1997a).

The role of cattle in deforestation varies, depending upon the type of production system. For example, large-scale “investment ranchers” (urban entrepreneurs with little tradition in cattle ranching) utilize cattle as a financial investment selected from their available alternatives (Kaimowitz, 1995a). A generic list of factors underlying the role of cattle in deforestation includes (Vosti and Witcover, 1996; Reardon and Vosti, 1995; Kaimowitz, 1995a, b; Hecht, 1993):

Financial incentives

Government subsidies provided to large-scale ranchers are usually identified as the primary ‘culprit’ underlying the rapid increase in the cattle herd of the Amazon (Fearnside, 1987, 1989; Hecht et al., 1988). Beginning in 1966, SUDAM (the Brazilian Superintendent for the Development of the Amazon) provided investment funds and tax benefits to ranchers. Normally, projects were organized as shareholding societies in which the participants were entitled to direct financial support, exemptions or reductions on international duties, exemptions and/or reductions on income taxes, and tax reinvestment benefits (Yokomizo, 1989). As of March 1989, cattle ranching accounted for 58 percent of the total number of approved projects, mostly in Mato Grosso and Pará.

SUDAM classifies approved projects into different categories: (1) implemented, (2) cancelled, (3) forfeit and (4) being implemented. As of 1988, only 15 percent of all cattle ranching projects approved in Legal Amazon had been fully implemented even though close to half of them (312 projects) were approved prior to 1973. Most projects (73 percent) were listed as ‘being implemented’ in 1988. Many of the ranching projects ‘being implemented’ had been in this category for over 10 years. In the 1980s a renewal of public efforts to stimulate cattle production in the Amazon caused almost 400 projects to be approved, including 229 in 1982 – 1984 alone. It is in this period that the mad rush to promote investment in the Amazon led to wide scale project approvals, even when the ranching projects had little chance of economic viability or when resources were not sufficient to fulfil commitments (Yokomizo, 1989; Hecht, 1985).

Financial incentives were expanded, supervision and control were relaxed and a carnival atmosphere emerged, where the chances of windfall gains of huge tracts of land and public subsidies caused investors to ignore the limited economic potential of many of the projects. The rapid expansion of project approvals in the early 1980s was matched by a decline in available resources. This can be seen in Figure 4 which shows cattle ranching project approvals on the right-hand axis and the index of available fiscal resources (in constant terms) on the left-hand axis, both from 1975 to 1985. The rapid growth in project approvals between 1982 and 1984 is matched by a significant decrease in resources for fiscal incentives. Not only was less money being spread among more projects, but serious and widespread administrative problems further reduced the effectiveness of the subsidy packages (Yokomizo, 1989). In the late 1980s subsidies for new projects were suspended, a moratorium which has continued until today.

Figure 4: SUDAM Supported Ranches - Number Approved and Resources

Figure 4

Source: Yokomizo (1989).

Land speculation

The spread of cattle ranching in the Amazon is usually tied to land speculation (Fearnside, 1989; Hecht, 1985, 1993; Browder, 1988; Barbier et al., 1991). The primary body of empirical evidence used to support this notion boils down to one set of data taken from Mahar (1979) and reproduced in graphical form by Hecht (1985), showing nominal land prices and the general consumer price index between 1966 and 1975. Combined with anecdotes (see review in Schneider, 1995), this one sample of data is essentially the sole empirical base used to support the claims of land speculation (e.g., see Schmink and Wood, 1992, p.2; Fearnside, 1990).

Other available data regarding Amazon land prices reveals that the potential for speculatory profits through the 1970s and 1980s was limited (Homma et al., 1991; Reydon and Herbers, 1989; Faminow, forthcoming). Relative to the rest of Brazil, real farmland and pasture prices in the Amazon were stagnant, reducing the returns from speculation. Actual rates of return that could be earned from general Amazon land price changes were quite low over the 1970 to 1988 period. The idea that broad and consistent land price increases have generally provided large speculative earnings to anyone who could buy land and hold on to is mostly a myth. Some investors/speculators earned handsome rates of return from buying and selling land, but this is not the widespread and pervasive phenomena that is commonly believed.

Property rights

In the Brazilian Amazon, serious property rights problems have been linked to deforestation (Alston et al., 1994). The Brazilian Government, along with individual state governments, owns most of the Amazon and has been the primary actor in land allocation. There are three primary categories of private property in Brazil. Individuals may hold definitive titles that are issued by Federal Government agencies. These titles can be defended in court, freely transferred and are recognized nationwide. An alternative ‘semi-official’ form of title is an authorization to occupy land, which is issued by the government and formally allows an individual to occupy land while waiting for the processing of a definitive title. Finally, squatters occupy public or private land without any formal documentation. Squatters have played a very significant role in the occupation of the Amazon because the assignment and transfer of property rights in Brazil is complex and the bureaucratic process unacceptably slow (Schmink and Wood, 1992).

Because of the slow-moving legal process, informal measures have been devised to allow farmers access to certain benefits normally associated with legal property rights. For example, a letter from an extension agent identifying an individual as a land occupant is accepted by banks when allocating official farm credit.

Brazilian law favours squatters who can demonstrate that they have occupied and improved land (Uhl et al., 1994). Because forested land has long been considered idle under Brazilian law, squatters have a clear incentive to deforest land quickly in order to demonstrate use and improvement, and thereby gain informal land rights and eventually a definitive title to the land. Definitive titles increase wealth through a rise in the value of the land caused by the ease of transferability and state-guaranteed property security that ensues. Note that large and small land owners may face the same incentive. By clearing forests they can assert their property rights and pre-empt squatters who might be tempted to settle on ‘idle’ forest land and make claims that could be granted under Brazilian law. Even in cases where land owners have secure title, there may be additional tax incentives to deforest land (Almedia and Uhl, 1995a).

The regional demand for cattle products

The role that regional demand for beef and dairy products has played in explaining the explosive growth of the cattle herd in the Amazon is overlooked in the literature. Population growth is generally seen as an influence upon deforestation rates, through supply-side influences such as clearance for agriculture production and the critical link to product consumption is generally not made. Anecdotal evidence suggests that much of the increase in the cattle herd in the Amazon region was demand driven (Homma et al., 1983; Resende and Gonçalves, 1985; Roberts, 1996). Colonization and resulting rapid population growth created expanding markets for livestock products that were imperfectly integrated to the rest of Brazil through a costly and overstressed land transportation network and also poorly integrated to external markets by trade policies that promoted self-reliance. The increase in the cattle herd closely followed the increase in beef demand from 1960 to 1980, but then grew much more quickly between 1980 and 1991. Population growth in the Amazon depended heavily upon the construction of highways that provide links and access to the rest of Brazil: (1) the Belém to Brasilia highway (north-south); (2) the Transamazonia highway (east-west) through the heart of the Amazon; and (3) the construction of BR-364 (north-south) linking Acre and Rondônia to the south. The long and costly transportation links provided a spatial market niche for cattle production at the fringe of service areas from other regions in Brazil (Faminow, 1997a).

Table 1 provides rough estimates of beef production for three years (1960, 1980 and 1991) and puts these data into perspective by showing the self-reliance percentage, calculated as beef production using off-take rates relative to beef consumption. Self-reliance calculations provide a very general view of production and consumption by abstracting away from regional trade possibilities, which can be problematic for a region the size of Legal Amazon, so care should be exercised in interpreting these data. By 1991, substantial changes had occurred. The cattle herd in Rondônia grew from trivial levels to the second largest in the region and was producing almost twice as much beef as was needed for consumption within the state. The cattle herd in Pará grew dramatically, but so did consumption so that the state has a relatively small surplus available for export to other states. Despite a huge increase in beef consumption, Amazonas had only a small increase in beef production and it remained a major deficit region, providing only a quarter of the beef consumed. Overall, the North Region was roughly self-sufficient by 1991 after having a self-reliance rate of 50-60% through the 1960 to 1980 period. Trade within the Amazon and also with other regions has increased over time.

Table 1: Beef Production and Self-Reliance in Brazilian Amazon states

 Beef Production (MT)Percent Self-Reliance(%)

Note: Estimates of beef production are made assuming different off-take rates from herd inventories for the three years. In 1960 it was assumed that the herds were reasonably stable but with the low off-take rate of 13%. Because 1980 was in the period of rapid herd expansion when a large proportion of heifers were being retained the very low off-take rate of 10% is used. Finally, the estimated off-take rate for 1991 of 15% was used.

Cattle management for better outcomes

Cattle production in the Amazon is decried as a low-productivity enterprise that is not capable of generating positive returns without subsidies and profits from land speculation. Yet, only about 500 of the over 50,000 medium to large-scale ranchers in the Brazilian Amazon ever received a generous incentive package from SUDAM. Cost engineering data from the mid-1980s support the nation that cattle ranching is unprofitable (Hecht et al., 1988; Browder, 1988), but data from the early 1990s suggest otherwise (Mattos and Uhl, 1994; Arima and Uhl, 1995; Vosti et al., 1997). Overall, revealed preferences and the extensive experience with cattle by large and small producers across the entire Amazon suggest that cattle can be a profitable enterprise (Faminow, 1997b).

What opportunities exist to improve cattle management (micro and macro level) in the agroecosystems being created in the Amazon? As a conceptual basis for assessing agroecosystem performance and choosing better management systems we suggest using the concept of the “critical triangle” linking sustainability, growth and poverty alleviation (Vosti and Reardon, 1997). By micro level we mean decisions which may be taken at the level of the household livelihood system or below. Macro-level decisions are taken at higher levels in the hierarchy through policy levers which influence cattle decisions in household livelihood systems.

Micro-level decisions

Productivity targets based upon intensive systems developed from agricultural research may be unrealistic for large-scale production in tropical regions because experiment station results do not usually reflect actual production levels in commercially-run farms (Nicholson et al., 1995; Preston and Leng, 1987). Farmers often resist adoption of animal and plant varieties or management systems that are developed in agricultural experiment stations (Reátegui et al., 1995). For example, annual beef production under experiment station conditions can range between 438 and 657 kilogram per hectare, four to six times above typical beef production rates in the region (da Costa et al., 1995; Mattos and Uhl, 1994; Almeida and Uhl, 1995b). Typical on-ranch beef production rates are usually under 150 kg/ha, and often a low as 50 kg/ha, when all deforested land (active pasture and fallow) is included in the denominator (Mattos and Uhl, 1994).

Achievement of productivity levels similar to the experiment station results usually requires that an entire technological package be adopted (good cattle genotype, proper nutrition, appropriate health management and reproductive control and management). However, because of scarce capital, risk considerations, and other economic criteria cattle producers might follow a step-wise adoption process and technology does not always “trickle down” (Nagy and Sanders, 1990; Wadsworth, 1990). Also, livestock production technologies may not match up with the objectives of producers (Preston and Leng, 1987; Nicholson et al., 1995). For example, smallholders who specialize in dual-purpose milk and beef production will need to trade off some desired characteristics (e.g., milk production) in order to get other characteristics (e.g., high beef production).

It is now widely agreed that, in order to be successful, agricultural technology must fit within the family farm livelihood system (Reardon and Vosti, 1995). Because the objectives for a farm family will necessarily be broader than those for the livestock system, application of narrow herd performance indicators will not always be an accurate indicator of whether a technology package will be adopted. Cattle represent one of the few viable alternatives for diversification and asset growth in the Amazon (Mattos and Uhl, 1994). Family wealth increases from growth in herd size and from increases in animal values can be a very important factor in farm decisions. For example, Léna (1991) provides data showing that accumulation of wealth through cattle herd ownership provides a vehicle by which departing family members can be rewarded for their contributions to the farm. A related objective is financial liquidity. For households without access to other financial investment instruments and formal insurance programs, cattle provide a means for liquidity for contingencies such as illness. Finally, cattle production may provide non-consumptive benefits such as work satisfaction and prestige, while oxen may be utilized for draft purposes and cattle integrated into crop production by using manure to fertilize soil (Arima and Uhl. 1995).

Macro-level decisions

Macro policy changes affect agricultural and non-agricultural sector activities, although in some cases pulling a macro/sectoral policy lever might have greater/less impact on livestock production systems than other forms of production. Product and production system characteristics offer clues as to the direction and relative magnitude of the impacts of policy change on cattle and other sub-sectoral activities. For example, for smallholder dairy herds without access to milk pick-up by dairy processors (a common case), many policies will have minimal effects because production is weakly integrated into the cash economy. Some factors influencing technology adoption are quite ‘macro’ in nature (e.g., exchange rates, interest rates, relative prices of broadly traded commodities, etc.) and are by definition beyond the control of local, state and sometimes even national policymakers, at least in the long term.

In our view, the most important macroeconomic process in the western Amazon has been and will continue to be regional integration. This regional integration is driven by investments in physical infrastructure and reductions in obstacles to trade, which decrease transactions costs and increase trade flows-with both existing trading partners and new trading partners, and for products currently traded and new ones. Several examples currently exist in the western Amazon of this phenomenon. First, within-state integration is already underway. For example, the state of Rondônia is investing heavily in rural infrastructure in order to link forest margin areas with the major market center of Porto Velho, and with secondary cities such as Ji-Paraná. Second, inter-state integration is increasing from infrastructure investments such as the paving of the Acre/ Rondônia segment of BR-364 which facilitated flows of inputs and outputs between these states. Finally, and perhaps most importantly, are the intra-regional integration efforts. Currently six “avenues to the Pacific” are on the drawing board or in the process of being constructed, including two which link Acre/Rondônia with Bolivia/Peru. Another example is the massive grain (and other products) transfer/transport facility recently inaugurated in Rondônia, which will integrate the major soyabean production region of the northern savannah in Brazil with regional and global markets. Given the northward migration of the Brazilian soyabean complex over time, it is an open question whether the opening of this facility will encourage the continued expansion of soybean into the southern fringes of the Amazon, possibly displacing pasture and shifting the cattle sector northward also. Moreover, the methods used to promote regional integration and the geographic areas included in the process can have major implications for production/consumption patterns. For example, recent investments in river transport signal a departure from the focus on (and massive investments in) overland transportation, especially more traditional routes linking north and south. This geographic reshuffling and route refocusing may have large implications for the geographic distribution of cattle production and transport.

The pace and direction of regional integration will by guided by comparative advantage in production and intra-regional differences in levels and patterns of consumption. Most important, with regional integration the boundaries of private profitability of production and trade will expand dramatically, jumping beyond state and national boundaries to incorporate new lands and consumers, thereby eliminating the rough supply-demand balance described in the earlier section of this paper. Similarly, north-south links within Brazil continue to improve. In light of MERCOSUL, the free trade agreement linking Brazil, Argentina, Paraguay, Uruguay and Chile (with probable expansion to other countries such as Bolivia, Peru and Columbia) increased regional integration is a virtual certainty.

Very imperfect financial markets and macro-economic instability have contributed to the use of cattle as an inflation hedge. The annual inflation levels experienced in Brazil over the past several decades are mind boggling, ranging from a low of about 15% in 1995 to a high of over 2,700% in 1993 (an average of over 500% over the entire period). For smallholders, cattle have offered one of the few viable alternative uses of surplus funds for protection from inflation and investment potential (Hecht, 1993; Mattos and Uhl, 1994). Improvements in financial markets and reductions in inflation (the two may be linked and, perhaps, linked in different ways in urban versus rural areas) will have an impact on the demand for cattle at the level of small rural households. The appearance of more efficient capital markets in rural areas should reduce the demand for cattle (on the hoof) as stores of wealth and as a medium of exchange—at a minimum because some portfolio shifting out of cattle and into cash balances will occur for pure risk-spreading reasons, and probably also because cattle might not be an efficient mechanism for doing so.

In Brazil, production taxes are levied at points of sale through the marketing chain and affixed to the sales price. Cattle producers can avoid paying the tax by selling cattle to small uninspected slaughtering plants. Official estimates of national slaughter are badly distorted because of the large clandestine slaughter market. For example, sources such as hide tanning receipts suggest that clandestine slaughter typically accounts for about 45% of all slaughter in Brazil and best guesses are that, currently, over 50% of all slaughtering (in Acre/Rondônia) is done illegally. The Brazilian government is making a concerted effort to reign-in this clandestine market with measures such as the strict enforcement of meat inspection regulations(e.g., by requiring all sales of meat to supermarkets to be in vacuum packages, a move which could curtail some output of the small clandestine packing facilities). If this initiative is successful, it should shift supply curves for beef inward by the amount of the tax levy.

Finally, serious efforts are being made to eradicate foot and mouth disease in Brazil through a nation-wide and highly visible campaign organized by the federal government in conjunction with individual state governments. Some major cattle producing states in the south and centre-west of Brazil(such as Rio Grande do Sul and Mato Grosso) have been free of foot and mouth disease for more than two years. This awareness and vaccination campaign has even spread to smallholder cattle producers in remote regions in Acre and Rondônia, where vaccination of herds against foot and mouth disease is an increasingly widespread and regular management practice because of extension efforts and more stringent testing by commercial dairies in the region.

A final word

Ultimately, one must question whether the best use of the Amazon rainforest is pasture for the production of beef and milk. The revealed preferences of large and small farmers and ranchers in the Amazon, along with on-the-shelf cattle production technologies and limited current markets for alternative crops, support the notion that cattle will continue to be the most widespread agricultural use of the region, at least over the next several decades.

From an environmental perspective expansion of this land use is not appealing. The rainforest provides a range of ecological services and potential economic products that are lost when converted to pasture. Protection of the remaining forest will depend upon four primary factors: (1) identifying and promoting alternative income-generating activities for rural households, especially those making use of already-deforested and, possibly, degraded lands; (2) the adoption of improved technologies for raising cattle by the large, medium and small landholders already in the Amazon; (3) the extent to which population and income growth in the Amazon continues; and (4) the capacity of governments as owners and managers of vast tracts of the Amazon to preserve the forest. The current trends to enhanced regional integration, along with increased international links, means that serious efforts along all four of these factors will be necessary for forest preservation.

Ultimately, it is the land-use decisions of the hundreds of thousands of households that currently occupy and raise cattle in the rural areas of the Amazon that will decide the fate of the forest. Policies can be taken at the micro and macro level to influence these decisions. The need to achieve growth and poverty objectives while eventually translating environmental trade-offs into complementarities suggests an important role for agricultural research. Adoption of improved technologies to intensify production and increase the effective carrying capacity of pastures would help provide a means for limiting future use of the forest for new pasture, relative to the extensive and, often, extractive systems widely used. The intensification option is not often supported by environmentalists but is one that surely deserves serious consideration.


Almeida, O.T. de and C. Uhl. 1995a. “Brazil's Rural Land Tax.” Land Use Policy 12 (No.2): 105–114.

Almeida, O.T. de and C. Uhl. 1995b. “Developing a Quantitative Framework to Sustainable Resource-Use Planning in the Brazilian Amazon.” World Development 23 (No. 10): 1745–1764.

Alston, L.J., G.D. Libecap and R. Schneider. 1996. The Determinants and Impacts of Property Rights: Land Titles on the Brazilian Frontier. NBER Working Paper Series 5405. Cambridge, MA. June.

Arima, E.Y. and C. Uhl. 1995. “Ranching in the Brazilian Amazon in a National Context: Economics, Policy and Practice.” Mimo. IMAZON. Belém.

Barbier, E.B., J.C. Burgess and A. Markandya. 1991. “The Economics of Tropical Deforestation”. Ambio 20 (No. 2, Apr.): 55–58.

Browder, J.O. 1988. “Public Policy and Deforestation in the Brazilian Amazon.” IN: Repetto, R. and M. Gillis. Public Policies and the Misuse of Forest Resources. Cambridge University Press. pp. 247–297.

Colinvaux, P.A. 1989. “The Past and Future Amazon.” Scientific American 260 (No. 5, May): 68–74.

Costa, N.A. da, L.O.D. de Moura Carvalho, J.F.T. Neto, J.B. da Veiga and W.S. Couto. 1995. “Sistemas de Pastejo Intensivo Rotacionado.” Pamphlet: Dia de Campo. Centro de Pesquisa Agroflorestal de Amazônia Oriental. EMBRAPA. Belém, Pará.

Faminow, M.D. 1997a. “Spatial Economics of Local Demand for Cattle Products in Amazon Development” Agriculture, Ecosystems and Environment 62 (March): 1–11.

Faminow, M.D. 1997b. “The Brazilian Cattle Sector: Status, Prospects and Controversies” Canadian Journal of Agriculture, Ecosystems and Environment 62 (March): 1–11.

Faminow, M.D. Forthcoming. Cattle, Deforestation and Development of the Amazon: An Agronomic, Economic and Environmental Perspective. Wallingford: CAB International.

Fearnside, P.M. 1983. “Land-use Trends in the Brazilian Amazon Region as Factors in Accelerating Deforestation.” Environmental Conservation 10 (No. 2, Summer): 141–148.

Fearnside, P.M. 1987. “Causes of Deforestation in the Brazilian Amazon.” In: Dickinson, R.E. (Ed.), The Geophysiology of Amazomia: Vegetation and Climatic Interactions. John Wiley and Sons: New York: pp. 37–53.

Fearnside, P.M. 1989. “Deforestation and Agricultural Development in Brazilian Amazonia.” Interciencia 14(No. 6, Nov./Dec.): 291–297.

Fearnside, P.M. 1990. “Predominant Land Uses in Brazilian Amazonia.” In: Anderson, A.B. (ed.) Alternatives to Deforestation: Steps Toward Sustainable Uses of the Amazon Rain Forest. New York: Oxford University Press: pp. 233–251.

Fearnside, P.M. 1993. “Deforestation in Brazilian Amazonia: The Effect of Population and Land Tenure.” Ambio 22 (No.8, December): 537–545.

Fearnside, P.M. 1997. “Environmental Services as a Strategy for Sustainable Development in Rural Amazonia.” Ecological Economics 20 (No. 1, January): 53–70.

Haller, A., R.S. Torrecilha, M.C. del P. da Haller and M.M. Tourinho. 1996. “The Socioeconomic Development Levels of the People of Amazonia Brazil - 1970 and 1980.” Journal of Developing Areas 30 (April): 293–316.

Hecht, S. B. 1985. “Environment, Development and Politics: Capital Accumulation and the Livestock Sector in Eastern Amazonia”. World Development 13 (No.6): 663–684.

Hecht, S. B., R.B. Norgaard and G. Possio.1988. “The Economics of Cattle Ranching in Eastern Amazonia.” Interciencia 13 (No. 5, Sep./Oct.): 233–240.

Hecht, S. B. 1993. “The Logic of Livestock and Deforestation in Amazonia.” BioScience 43 (No. 10): 687–695.

Homma, A. K. O., P. C. Kitamura and G. H. H. Flohrshütz. 1983. Análise do Complexo Pecuária no Nordeste Paranense EMBRAPA-CPATU. Documentos, 18. Belém, Brazil.

Homma, A. K. O., A. R. T. Filho and E. P. Magalhães.1991. “Análise do Preço da Terra como Recurso Natural Durável: O Caso da Amazônia.” Revista de Economia e Sociologia Rural 29 (No. 2, April/June): 95–102.

Homma, A. K. O., R. T. Walker, F.N.Scatena, Conto, R. de A. Carvalho, C.A.P. Ferreira and A. I. M. dos Santos. 1995. “Redução dos Desmatamentos na Amazônia: Política Agrícola ou Ambiental?” Paper presented at the XXXIII Congresso Brasileiro de Economia e Sociologia Rural. July 31 - August 3.Curitiba, Paraná.

Kaimowitz, D. 1995a. “Livestock and Deforestation in Central America in the 1980s and 1990s: A Policy Perspective.” EPTD Disscussion Paper No. 9 (June). International Food Policy Research Institute and Interamerican Institute for Cooperation on Agriculture: San Jose, Costa Rica.

Kaimowitz, D. 1995b. “Land, Tenure, Land Markets and Natural Resource Management by Large Landowners in the Petén and the Northern Transversal of Guatemala.” Paper Presented at the 1995 Meeting of the Latin American Studies Association, The Sheraton, Washington, September 28–30.

Léna, P.1991. “Ritimos e Estratégias de Acumulação Camponesa em Áreas de Colonização: Um Exemplo em Rndônia.” Boletim Museu Paraense Emílio Goeldi, Sér Antropologia 7 (No. 1): 37–70.

Mahar, D. 1979. Frontier Policy in Brazil: A Study of Amazonia. New York: Praeger.

Mattos, M.M. and C.Uhl. 1994. “Economic and Ecological perspectives on Ranching in Eastern Amazon.” World Development 22 (No.2): 145–158.

Moran, E. 1993. Through Amazon Eyes: The Human Ecology of Amazonian Populations. lowa City:University of lowa Press.

Myers, N. 1993. “Tropical Forests: the Main Deforestation Fronts.” Environmental Conservation 20 (No.1, Spring): 9–16.

Nagy, J.G. and J.H. Sanders. 1990. “Agricultural Technology Development and Dissemination Within a Farming System Perspective.” Agricultural Systems 32 (No. 4): 305–320.

Nicholson, C. F., R. W. Blake, C. I. Urbina and D. R. Lee. 1995. “Livestock, Deforestation and Policy Making: Intensification of Cattle Production Systems in Central America Revisited.” Journal of Dairy Science 78 (No.3, March): 719–734.

Preston, T. R. and R. A. Leng. 1987. Matching Ruminant Production Systems With Available Resources in the Tropics and Sub-Tropics. Armidale: Penambul Books.

Psacharopoulos, G., S. Morley, A. Fiszbein, H. Lee and W. C. Ward. 1995. “Poverty and Income Inequality in Latin America During the 1980s.” Review of Income and Wealth 41 (No.3, September): 245–264.

Reardon, T. and S. A. Vosti. 1995. “Links Between Rural Poverty and the Environment in Developing Countries: Asset Categories and the Investment Poverty.” World Development 23 (No.9, September): 1495–1506.

Reátegui, K., R. R. Vera, W. L. Loker and M. Vásquez. 1995. “On Farm Grass-Legume Pasture Performance in the Peruvian Rainforest.” Experimental Agriculture 31: 227–239.

Resi, E. J. and S. Margulis. 1991. “Options for Slowing Amazon Jungle Clearing.” In: Dornbusch, R. and J. Poterbar (eds.) Economic Policy Response to Global Warming. Cambridge: MIT Press.

Repetto, R. 1990. “Deforestation in the Tropics.” Scientific American 262 (No.4, April): 36–42.

Resende, J. C. de and C. A. Gonçalves. 1985. Pecuária Bovina no Estado do Rondônia: Dagnostico e Aspectos Econômicas. EMBRAPA-UEPAE.Documentos, 15. Porto Velho, Brazil.

Reydon, B. P. and R. G. Herbers. 1989. “Politica Governmental Para a Agropecuária e Degredaçã do Meio Ambiente.” Reforma Agrária (April/June): 42–51.

Roberts, J. J. 1995. “Trickling Down and Scrambling Up: The Informal Sector, Food Provisioning and Local Benefits of the Carajas Mining Growth Pole.” World Development 23 (No. 3): 385–400.

Schmink, M. and C. H. Wood. 1992. Contested Frontiers in Amazonia. New York: Columbia University Press.

Schneider, R. R. 1995. Government and the Economy on the Amazon Frontier. World Bank Environment Paper No. 11. Washington.

Skole, D. and C. Tucker. 1993. “Tropical Deforestation and Habitat Fragmentation in the Amazon: Satellite Data from 1978 to 1988.” Science 260 (June 25): 1905–1910.

Uhl, C., A. Verissimo, P. Barreto, M. Mattos and R. Tanita. 1994. “Lessons from the Aging Amazon Frontier: Opportunities for Genuine Development. In: Kim, K. C. and R. D. Weaver (eds.) Biodiversity and Landscape: A Paradox of Humanity. Cambridge: Cambridge University Press.pp. 287–303.

Villachica, H., J. E. Silva, J. R. Peres and C. M. C. da Rocha. 1990. “Sustainable Agricultural Systems in the Humid Tropics of South America.” In: Edwards, C. A., R. Lal, P. Madden, R.H. Miller and G. House (eds.). Sustainable Agricultural Systems. lowa: Soil and Water Conservation Society: p. 391–437.

Vosti, S. A. and J. Witcover. 1996. “Slash and Burn Agriculture -- Household Perspectives.” Agriculture, Ecosystems and Environment 58: 23–38.

Vosti, S. A., and T. Reardon. 1997. “1. Introduction: The Critical Triangle of Links among Sustainability, Growth. and Poverty Alleviation.” In: Vosti., S. and T. Reardon (eds.). Sustainability, Growth, and Poverty Alleviation: A Policy and Agroecological Perspective. Baltimore, MD: John Hopkins University Press for the International Food Policy Research Institute.

Vosti, S.A., S. Oliveira and M. Faminow. 1997. “Policy Issues In Agroforestry: Technological Adoption and Regional Integration in the Western Brazilian Amazon. Paper presented at the 88th Tri-Societies Meeting. February. Indianapolis.

Wadsworth, J. 1990. “Developing Efficient Extension Strategies: Results of an Experiment Involving Costa Rican Livestock Producers.” Agricultural Systems 32 (No.3): 259–275.

Yokomizo, C. 1989. “Financial and Fiscal Incentives in the Amazon.” In: Universidade de São Paulo. Amazônia Facts Problems and Solutions. São José dos Campos, São Paulo: pp. 93–136.