Evaluation of Conservation Policies in the Brazilian Amazon
No Thumbnail Available
Files
Date
2010-04-30
Authors
Journal Title
Series/Report No.
Journal ISSN
Volume Title
Publisher
Abstract
The debate about how to conserve nature while promoting sustainable development is ongoing and seemingly never-ending. A key question in this debate is how feasible is it to integrate conservation and people in tropical forest regions. The debate over this approach has been acrimonious and characterized by sweeping generalizations about the negative effects of protected areas on local people, the complete failure of integrated conservation and development, and the lack of sustainability of externally funded projects. In this dissertation, I examine some of these claims by applying rigorous empirical methods to rich panel datasets that I have compiled for this purpose, seeking lessons from the nuances and variation in outcomes, with the goal of moving beyond the generalizations.
The three essays in this dissertation focus on different dimensions of the relationship between protected areas (PAs) and human welfare in the Brazilian Amazon, which has the largest contiguous remaining area of tropical forest in the world, half in protected areas. First, I evaluate the impact of PAs on the incidence of malaria, a reemerging and endemic disease with 1.9 million annual cases in the Brazilian Amazon. Malaria is prevalent in disturbed forest environments and its association with deforestation has been well documented. My contribution is to assess the link between malaria and policy variables, specifically different types of PAs and roads. Since some of the key policies of interest do not vary across time in my dataset, I use an innovative fixed effects vector decomposition method to include these time-invariant variables in a panel model. I find that sustainable use PAs have the unintended consequence of increasing malaria prevalence, while strict protection PAs and indigenous reserves do not have this effect.
The second and third essays focus on a particular sustainable use protected area, the Tapajós National Forest. Specifically, I evaluate participation in and effects of community-based microenterprises supported through an ICDP (Integrated Conservation and Development Program). Chapter 2 evaluates the program in terms of the impacts of participation during the external financing phase using a panel dataset formed by household observations in 1997 (2 years before the ICDP was implemented) and 2006 (the year funding ended). I control for the potential confounding effects of program placement and selection by applying propensity score matching to evaluate difference-in-difference outcomes. Participation in the microenterprises results in higher cash income, but this is not reflected in greater asset wealth or other indicators of well-being. Participation also not have consistent effects on agricultural cropping area and size of the cattle herd, the direct environmental targets of the ICDP. Considering that $1.4 million was invested in the ICDP, the lack of impact at the household level is a significant finding.
In my third essay, I evaluate the retention of households in the microenterprises and decisions about agricultural cropping area in the Tapajós National Forest between 2006 and 2008, when operation of the microenterprises was transferred to the communities. I hypothesize that social networks have a significant role in determining who stays in the projects and who decreases land in agriculture (two measures of success relevant to the external funders). Estimation results show that social networks do have an important influence, and that the different microenterprises have heterogeneous effects (with some more likely to retain household participants, and some more likely to encourage reductions in agricultural areas). Overall, my results offer cautionary lessons about the difficulty of conserving nature while promoting sustainable development and illustrate the importance and need for well-designed empirical research.
Description
Keywords
Amazon, program evaluation, land use, malaria, deforestation, Time Series Cross Section, social networks, microenterprise
Citation
Degree
PhD
Discipline
Forestry
