Bureaucracy matters: Administrative structure and performance in Brazil’s federal protected areas agency

Environmental agencies in developing countries frequently struggle to implement policy. Yet few researchers seeking to explain environmental outcomes in the developing world have focused on the internal functioning of environmental agencies. I address this gap by analyzing the relationship between administrative structure and regulatory effectiveness in Brazil’s Chico Mendes Institute (CMI), a federal agency which oversees one of the world’s largest systems of protected areas. I focus on two questions. First, how might the re-allocation of CMI's personnel improve its performance? Building on findings suggesting the potential for substantial gains from personnel re-allocation, I then ask: what explains the severe misallocation observed? To answer question (1), I use fixed effects OLS regression to estimate how the effect of protected area management team size on deforestation varies based on key predictors of deforestation pressure. This analysis covers 322 of the agency’s 334 protected areas over the first ten years of CMI's existence (2008-2017). Model results suggest that various re-allocation initiatives may have prevented roughly 2,200 km2 of deforestation—approximately 38 times the area of Manhattan—over the study period. More than 40 interviews with CMI staff who worked in the Amazon region during this period complement the econometric analysis. To understand what accounts for the severe workforce imbalance observed, I primarily rely on interviews (n=45) and process tracing. In addition to the field staff mentioned above, informants included former agency executives and headquarters-level staff who are familiar with personnel allocation decision-making processes. Results suggest that the imbalance results from multiple causes. I highlight the role of two institutions in preventing the agency from addressing rangers’ preferences for working in biomes facing greater deforestation risk. First, I emphasize resistance by the Ministry of Planning to the installation of incentive-based personnel management systems commonly used to prevent workforce imbalances. Second, I underscore labor laws, which limit the agency’s ability to mandate transfers. These findings carry important implications for environmental governance scholarship focused on developing countries, which tends to focus on political obstacles to policy adoption and policy implementation, as well as policy design. They demonstrate that administrative structure can greatly affect the extent to which a public environmental agency is able to meet its objectives. Furthermore, they illustrate how less commonly discussed political institutions—i.e. resistance from agencies responsible for governing a civil service and public service laws—can undermine environmental outcomes.