Agriculture is highly dependent on environment conditions, mainly temperature, precipitation and soil quality, thus it becomes the most vulnerable economic sector to the new climate conditions projected for the next decades. Therefore, knowing these impacts and the consequences for the rest of the economy is essential to map the effects and to elaborate, if necessary, mitigating environmental and economic policies. However, studies focusing on Brazil based on more regionalized data but linked to the rest of the world using dynamic computable general equilibrium (CGE) models are still very incipient. So this is precisely the gap that this article intends to fill, offering a modest contribution to the debate. Then, the objective of this paper is to determine the economic impact of the estimated changes in average agricultural productivity for the coming decades using a dynamic CGE model, the PAEGDyn linked to GTAP. Basically, the results found confirm trends in other works: the tropical regions in the world will be the most affected by the probable increase in the planet temperature, decreases in agricultural productivity and, thus, a reduction in economic performance.