Climate Risks and the Carbon Market in Brazil: A Study Using Computable General Equilibrium Models
Climate change. Carbon market. General equilibrium models. GHG emissions. Environmental policies.
The goal of this thesis project was to adapt a Computable General Equilibrium (CGE) model inspired by Caliendo, Dvorkin, and Parro (2019) to also address environmental issues. The original model by these authors is dynamic, features a broad spatial structure, and stands out for explicitly incorporating several relevant mechanisms: labor mobility frictions, trade costs, intersectoral input-output linkages, international and intersectoral trade (with the possibility of including interregional trade), and consumption of local final goods aggregated via a Cobb-Douglas function (with the possibility of extension to CES preferences). However, the main methodological contribution of the model by Caliendo, Dvorkin, and Parro (2019) is the use of dynamic exact-hat algebra, which enables solving for equilibrium and conducting counterfactual analyses without needing to estimate the levels of economic fundamentals (such as productivity, migration frictions, or trade costs). The adaptation proposed in this thesis consists of including emissions as a factor of production, following the approach of Copeland and Taylor (2013), within the original framework of Caliendo, Dvorkin, and Parro (2019). This theoretical extension allows for the analysis of environmental policy instruments such as carbon markets and Pigouvian taxes on emissions, within a consistent general equilibrium framework that incorporates trade and migration. We consider this theoretical extension to be the main contribution of this thesis project, as it enables an integrated evaluation of environmental and trade policies in a dynamic and spatially structured model. To ensure the project went beyond the theoretical realm, we propose a practical simulation of a carbon market in Brazil starting in 2025, with three different scenarios for reducing greenhouse gas (GHG) emissions by 2030, taking 2005 as the baseline year. All scenarios aim for a 50% total reduction in emissions by 2030, reaching approximately 1.28 Gt CO₂e, but they differ in how this target is distributed between deforestation and productive sectors. In this thesis, we completed the implementation and analysis of Scenario I (in which all sectors, including deforestation, reduce their emissions by 50%). The results show that implementing a carbon market to achieve the 2030 target leads to a significant initial contraction in the Brazilian economy (measured by Value Added – VA) and a slight decline in employment in the short term. Although VA levels remain below the baseline scenario, the growth rate recovers in the medium term, and employment levels tend to converge. The remaining scenarios (II and III) will be presented during the thesis defense, as their simulations are still in the process of being finalized and validated.