Emissions and Output: Evidence of Decoupling from Brazil

This paper provides evidence on the relationship between greenhouse gas (GHG) emissions and real GDP growth in Brazil, using both aggregate and state-level data. The focus is on distinguishing longer-run trends in this relationship from short-run cyclical fluctuations, and on documenting changes in these relationships over time. We use two different filtering techniques for separating trend and cycle, namely the commonly-used Hodrick-Prescott filter and the newer filter proposed by Hamilton (2018). We find that the long-run or trend elasticity for Brazil is about 0.8, higher than than in the average for advanced countries but below that of other major emerging markets. This elasticity is somewhat higher for consumption-based emissions than for production-based emissions. The direction of these results is not entirely consistent with the pollution haven hypothesis but the quantitative difference between the two elasticities is not large (or statistically different from one another). Consequently, our findings suggest that shifts in the allocation of global production across countries are not a major driver of shifts in the pattern of emissions growth in Brazil. Additional evidence comes from state-level data analysis where one can observe a great deal of heterogeneity, but also some hope as far as decoupling is concerned. Our analysis for 27 states does not reveal any underlying inverted U-shape relationship between the intensity of emissions-GDP relationship and the level of per capita GDP. That is, at the state level, we do not observe that Kuznets elasticity initially increases with provincial per capita real GDP but then declines. In addition to the trend relationship between emissions and output, we find that there does not seem to exist a short-run or cyclical relationship holding in Brazil at the aggregate level (despite having become more procyclical over time), but it does exist in a few individual states. Finally, we further find evidence of asymmetry in the cyclical relationship: the elasticity is higher in booms than during busts, and thus Brazil’s emissions go up more when GDP is above trend than they decrease when GDP is below trend. However, this result seems to be mainly driven by a few states. The transition to a low-carbon economy will thus require continued progress not only in bringing down trend emissions, but also in taming the increase in emissions that occurs during the boom phase of the business cycle.