Agricultural expansion and deforestation in lowland Bolivia: the import substitution versus the structural adjustment model

Abstract

This paper analyzes the influence that both policies and markets have on driving forest conversion to agricultural land use focusing on the Bolivian case. It assesses the effects that two models of economic development have had on prompting agricultural expansion, and hence deforestation in lowland Bolivia. The first model, of economic diversification and import substitution, initiated in the early 1950s, stimulated the occupation of forestlands for agricultural expansion through colonization, and incentives to developing a medium and large-scale agriculture. The second model of development, initiated in the mid-1980s, prompted by structural adjustment policies, led to an intense process of agrarian change in the lowlands, mainly by connecting the agricultural frontier to the international market of commodities, particularly soybeans. The move from an import substitution industrialization model to another based on economic liberalization, even though it had positive implications on stimulating agricultural growth, has led to accelerate the pace of deforestation in lowland Bolivia, which has a larger contribution of medium and large-scale agriculture. This paper moves forward the debate about deforestation in Bolivia by making explicit connections between some analytical assessments on land-use/cover change and empirical evidences about forest removal in a five-decades period.

Introduction

The debate on the dynamics of land-use/cover change, and specifically on the causes of deforestation, is progressively evolving toward higher levels of complexity in its attempt to explain the causal relationships influencing forest removal for other land uses (Geist and Lambin, 2001). There are two dominant perspectives: the first looks at the implications of household behavior on land-use decisions, and the second focuses on assessing the drivers of land use change at a broader level of understanding including shifts in policies and markets. It has been suggested that while the household socio-economic situation conditions the trajectories of deforestation, both policies and market conditions contribute to explain its magnitude (Brondizio et al., 2002; Moran et al., 2000).

This paper aligns with the second perspective. It assumes that some additional knowledge on causal relationships of drivers of land-use change can be gained by looking at both specific government policies and market conditions, the two of which determine what is labeled here a “model of regional development”. In this regard, Klepeis and Turner (2001) suggest that assessing different episodes of development, each shaped by specific policies and markets influencing agriculture expansion, provides a powerful instrument to understand the nature–society interactions for specific places through time.

It is very well known that policies and markets can have contradicting effects on deforestation because they can either prompt forest clearing for other land uses or favor forest conservation (Brown and Pearce, 1994; Contreras, 1999; Gillis and Repetto, 1988; Kaimowitz and Angelsen, 1998). The matter, therefore, consists of determining what are the outcomes resulting from specific policy and market situations, both determined historically. This paper does so through a case study in Bolivian lowlands,¹ where the rates of deforestation have tended to increase during the last five decades.

Two periods are analyzed corresponding to two different models of development, and within each of them two sub-periods of expansion and crises are identified. The first model, initiated in 1952, privileged policies seeking to diversify the economy by supporting the agricultural, mining, and industrial sectors, and substituting food imports by producing them internally within the country. The second model of development, initiated in 1985, based the growth of the national economy on a program of economic liberalization underpinned by structural adjustment policies.

In the early 1950s, the Bolivian government implemented policies favoring the expansion of the agricultural frontier through a dual scheme of frontier development. It supported the expansion of colonist settlements, as well as the modernization of medium- and large-scale agricultural establishments. Yet, the limited public investment to support agriculture and the reduced size of the domestic market were the main obstacles inhibiting the expansion of the lowlands’ agricultural frontier. High transportation costs along with trade barriers limited the expansion of agricultural exports. As a result, deforestation rates remained at comparatively low levels until the mid-1980s.

Since that time, the implementation of a national structural adjustment program largely stimulated the expansion of agricultural frontier in lowland Bolivia, mainly due to an increasing acreage of land devoted to soybean production, which is sold in the regional markets. Other factors favoring this expansion were the availability of cheap lands, fiscal incentives to stimulate non-traditional exports, as well as favorable market conditions in the Andean countries. Those conditions favored the agricultural frontier expansion leading to growing rates of deforestation to levels never reached before.

This paper analyzes the implications of policies on deforestation for the two periods described above. The main issues emphasized here are the in-migration trends to the lowlands, the market dynamics for agricultural products, and the shift from protectionist to neo-liberal economic policies. It also underscores the contribution of different agents to deforestation, viz, colonists, medium- and large-scale agricultural producers, and cattle ranchers, due to the political connotation that this issue has in the deforestation debate, and hence in the policies that are adopted to arrest deforestation.

As noted before, this paper argues that the structural adjustment policies led to larger rates of deforestation compared to protectionist policies aimed at substituting imports. This should not lead to mistaken interpretations suggesting that for deforestation to be reduced, protectionist policies should be reinstalled. It only attempts to assess the implications that different pathways of development had on deforestation.

Looking at the human–environment relationships over a long span of time demands long series of deforestation estimates. Two data sets have been merged to provide consistent evidences of deforestation covering the period analyzed here, even though both use different methodologies to estimate land-use change. The first data set, prepared by the Department of Geography of Maryland University (Steininger et al., 2000b), offers a three-period estimate of forest removal (until mid-1970s, from mid-1970s to mid-1980s, and from mid-1980s to mid-1990s), while the second data set, elaborated by BOLFOR, estimates it until 2000 (Rojas et al., 2003). Some adjustments have been made to the second one, to make the two data sets comparable, and to minimize potential errors.

This paper is organized in six parts including this introduction. The second section presents some of the theoretical debate supporting the relevance of considering policies and markets as fundamental drivers of forest clearing. The third section describes the forest in lowland Bolivia, as well as the land tenure situation, and provides some evidences of both magnitude and rates of deforestation in the last decades. The fourth section discusses the influences of both policies and markets on deforestation in the Bolivian lowlands for the import substitution period, and the next does the same for the structural adjustment period. The final section pulls together the main conclusions.