Testing the EKC hypothesis for the USA by avoiding aggregation bias: a microstudy by subsectors.

The aim of this study is to examine the impact of foreign direct investments, economic growth, and energy consumption on carbon dioxide subcomponents in the case of the USA. Dynamic ARDL (DARDL) econometric method is used covering the period 1972-2020. In addition to the total CO2 emission, the subcomponents of CO2 emission are examined separately within the framework of the EKC hypothesis in the USA by avoiding aggregation bias for the first time. The CO2 emission subcomponents used in the study are as follows; CO2 emissions from liquid fuel consumption, residential buildings, and commercial and public services; electricity and heat production; and other sectors, excluding residential buildings and commercial and public services, and CO2 emissions from transportation. Each CO2 emission component is used as a dependent variable and 6 different models were created. Foreign direct investments, trade, and energy consumption are used as control variables. No results supporting the EKC hypothesis are determined in any model, except for model 1, where total CO2 emission is the dependent variable. In addition, the trade variable has been determined as an important factor in reducing CO2 emissions in the short and long term. Trade and GDP per capita increasing and energy consumption reducing will show positive results in order to increase the environmental quality in the USA. Moreover, the study in which this EKC hypothesis is tested with CO2 emission and its subcomponents is an important study in terms of providing the opportunity to analyze the environmental quality from different angles at the same time and to take various measures together in the US economy.