Research Seminars @ Illinois

Abstract:
Brazil faces a fundamental tension between its commitments to reduce deforestation and its ambitions to consolidate its role as a global food supplier. This tension is especially acute in the cattle sector: the primary driver of deforestation in the Amazon, yet still heavily credit-constrained and dependent on rural credit, especially subsidized one. Rural credit is the Brazilian government's main agricultural policy instrument, yet little is understood about its role in deforestation. Existing evidence suggests that restricting credit access for non-compliant ranchers and farmers reduces deforestation in the Amazon. Still, there remains uncertainty about how non-excluded producers adjust their land-use and production decisions when they access credit. In this paper, I seek to address this gap. Using administrative microdata on loans from Brazil's Constitutional Financing Fund of the North (FNO), satellite-based land-use information, and cattle movement records for the state of Pará between 2013 and 2022, I construct a property-level panel that allows me to track how ranchers change their land-use and production decisions once they access subsidized FNO credit. I use a shift-share instrumental variables design that leverages annual fiscal transfers from the Brazilian National Treasury to the FNO, which are predetermined by a fixed constitutional rule. The analysis sheds light on how effective subsidized credit is in contributing to a sustainable transition in the Brazilian Amazon cattle sector and whether its design ultimately relieves or intensifies pressure on the forest.