Institutions and deforestation in the Brazilian amazon: a geographic regression discontinuity analysis

Abstract

This study explores the impact of institutional quality at the municipal level on deforestation in the Legal Amazon. We add to this insufficiently understood topic by implementing a geographic regression discontinuity design. By taking advantage of high-resolution spatial data on deforestation combined with an objective measure of corruption used as a proxy for institutional quality, we analyse 138 Brazilian municipalities in the period of 2002-2004. Our empirical findings show no causal effect of institutional quality on deforestation, suggesting that other unobserved factors are more important drivers of deforestation. A supplementary analysis indicates that location is a vital underlying factor in the interplay between deforestation and institutional quality. However, further research is needed to establish inference with respect to the relationship between deforestation, institutional quality and location.