Analysis of energy consumption on the environmental Kuznets curve hypothesis in the United States: does renewable energy play a role?

Abstract

Using CO₂ emissions as a representation of environmental degradation an empirical econometric analysis is conducted to see if there is evidence of an Environmental Kuznets Curve in the United States and if renewable energy consumption plays a significant role in CO₂ emission mitigation. The renewable energy consumption variable was broken down further to isolate geothermal, hydroelectric, biomass, solar, and wind energy consumption and explore their role in the analysis. An Auto-Regressive Distributed Lag approach to cointegration with Pooled Mean Groups and Mean Groups estimations was used on U.S. state (including District of Columbia) specific data from 1987 to 2015 to calculate the long and short run results that would support an Environmental Kuznets Curve hypothesis. The panel of states was divided into low, medium, and high GDP brackets as disaggregate models and those were examined along with a model of the entire United States. Evidence for an Environmental Kuznets Curve for the United States could not be established in the aggregate model, however it was found that renewable energy consumption did have a negative coefficient, which indicates CO₂ emission mitigation through renewable energy consumption. Out of the individual renewable energy consumption variables tested, only wind energy consumption was found to be statistically significant while the model also exhibited evidence to support an Environmental Kuznets Curve hypothesis in this aggregate model. Looking at the different GDP state brackets, low GDP states were the only bracket that yielded evidence of an Environmental Kuznets Curve in the disaggregate models. For estimations with the low GDP states bracket looking at the individual renewable energy consumption variables, hydroelectric, biomass, solar, and wind energy consumption variables were statistically significant as well. The medium GDP bracket states aggregate model did not yield conclusive results, stemming from the lack of slope in the GDP variable for this model. Out of the individual renewable energy consumption variables tested in the subset, biomass was the only energy consumption to be statistically significant while the model exhibited evidence of an Environmental Kuznets Curve. The high GDP bracket aggregate model did not yield results showing evidence of an Environmental Kuznets Curve, while the individual renewable energy consumption variable subset models geothermal and wind energy consumption were statistically significant within models showing evidence of an Environmental Kuznets Curve. Breaking out these separate renewable energy consumption variables in an Environmental Kuznets Curve analysis can provide empirical support for policy and investment in specific renewable energy technology.