Annual Report 2018

Economics CONSERVATION OF NATURE AND THE REACH OF THE STATE: EVIDENCE FROM NATIONAL FRONTIERS IN THE AMAZON Preservation of natural ecosystems in developing countries is a challenge. Because wilderness ecosystems are in remote location, the exercise of governmental control is hard. This study explores the degree to which the state can exert regulatory control over these remote wilderness areas by examining whether there are discrete changes in deforestation at the Brazilian international border in the Amazon region. OBJECTIVE • The aim is to investigate the reach of the State's power to impose conservation policies in remote areas, by evaluating data on forest clearance in regions around international borders in the Amazon. RESEARCH METHOD . • The analysis considered the entire Amazon region, according to the definition of the Amazon Network for Georeferenced Socio-environmental Information (RAISG), taking into consideration the limits of the biome and the "Legal Amazon" as defined by various countries in the region. A new dataset from Landsat 7, with annual resolution of 30 meters, was used to study the impact of the national policy on the Amazon by monitoring deforestation consistently over time (2000 to 2014) and space. The high resolution of these images allowed magnifying them in areas near the borders (the preferred specification uses a strip with width of 17 km on both sides of the border) to identify effects precisely. Spatial regression discontinuity (RD) designs were employed, using as variable the dis- tance to the Brazilian border, for forests where: (i) the portion of capital in production is low; (ii) the local supply of capital is probably elastic; and (iii) the basic production factor (land) is fixed in space, so the discontinuity in the regression would probably reveal the impact of the conservation policy, and the direct effects tended to predominate. Annual deforestation rates on both sides of the border were plotted between 2001 and 2014 and RD models were estimated separately for each year. 3. strengthened in 2006 by the Law on Management of Public Forests and the entry into operation of the Center for Environmental Monitoring, which together allowed the Brazilian State to combine detection of deforestation with police and military operations where illegal deforestation was detected. The legal restrictions on land use on the Brazilian side of the border made a differ- ence during the period studied. The areas designated as protected in Brazil always were less deforested than unprotected areas on the other side of the border, and this continued to be case form 2006 onward. The Brazilian State was therefore able to apply the environmental regulations when there was a desire to do so even in these peripheral areas. Combined, these results demonstrate a notable reach of the Brazilian State. They suggest that the accelerated deforestation in the Brazilian Amazon was partially was a conse- quence of lax enforcement of laws prior to the mid-2000s, and conversely, that the pre- cipitous decline in deforestation seen since the mid-2000s is the consequence of better monitoring and tighter enforcement. CONTRIBUTIONS OF THE STUDY • Using satellite images with fine resolution, the study tested whether Brazilian conserva- tion policies have had an effect along the nation's borders. This is an interesting exercise, because there is considerable skepticism in relation to the capability of the State to exer- cise control over remote regions. Indeed, the capability of the State to preserve ecosystems can decline as the areas be- come more remote, which opens opportunities to extract resources illegally. Given that the rapid environmental degradation in developing countries is being driven by illegal extraction, it is important to asses empirically whether or not the State has the power to conserve natural resources in these remote places. APPLICATIONS OF THE RESULTS AND POSSIBLE EXTENSIONS OF THE STUDY The fact that the deforestation rate in Brazil went from a maximum in 2004 to a minimum in 2009 bears witness that weak preservation policies can be strengthened. Part of this turnaround was achieved because the Brazilian State combined stricter environmental regulation with the use of satellite data in an integrated system involving various govern- mental entities: Brazilian Institute of the Environment and Renewable Natural Resources (IBAMA), Federal Police and the Army, among others. The success of wilderness conservation, therefore, ultimately depends on the policy choices of national governments. Information on illegal logging, for example, is regularly available to any government. The noteworthy reversion documented in the Brazilian Am- azon suggests it is possible to reduce deforestation and environmental degradation even in wild areas with difficult access in developing countries. RESULTS 1. The study documented three noteworthy facts: 2. Until 2005, the level and rate of deforestation were dramatically higher on the Bra- zilian side of the border than on the other side. In the starting year, 2000, Brazilian areas were about 30% more likely to be cleared than similar areas located nearby in neighboring countries. And from 2001 to 2005 the annual deforestation rate was three times higher on the Brazilian side of the border. The discontinuity in the deforestation rates fell abruptly in 2006 - when Brazil started to implement substantially stricter national policies to fight illegal deforestation. - as part of the Action Plan to Prevent and Control Deforestation in the Legal Amazon (PPCDAM), which increased the penalties for illegal deforestation, especially in un- claimed lands and private lands outside formally protected areas. The PPCDAM was 18 AUTHOR: Francisco J. M. da Costa. RESEARCHERS: Benjamin Olken and Robin Burgess. ORGANIZATION: EPGE Brazilian School of Economics and Finance (EPGE). SUPPORT: European Research Council. Annual Report 2018 19 RESEARCH

View entire presentation