Testing the Kuznets Curve Hypothesis at the Sector Level: Evidence from PNARDL for OECD Countries

Abstract

Since the contribution of economic sectors to GDP is heterogeneous, they also contribute differently to CO2 emissions and environmental degradation. Although many studies investigate the Environmental Kuznets Curve (EKC) hypothesis, there are still gaps in this literature, specifically in sectors. This study contributes to the literature in several ways. First, it examines the validity of the environmental EKC hypothesis for four different sectors (Agriculture, Industry, Manufacture and Service). Second, the impact of energy consumption on CO2 is examined asymmetrically. Finally, annual data from 38 OECD countries from 1990 to 2022 are used to fill this gap. The method used in the study is a new panel nonlinear ARDL to capture the asymmetric effect of energy consumption on CO2. The findings show that the EKC hypothesis is valid for sectors other than the industrial sector and for the overall economy. The turning point of the Kuznets Curve for the total economy is estimated as $29250 GDP per capita. The PNARDL model's error correction term (ec) is negative and significant. However, the magnitude of the term suggests that the system will return to long-term equilibrium after approximately 2.87 years following any shock. It has been determined that agriculture is the sector that causes the most environmental degradation by increasing CO2 emissions. This situation highlights the limited availability of productive capital equipment and renewable sources in the agricultural sector, even in OECD countries. Moreover, the empirical results clearly show that energy consumption has asymmetric effects on emissions in both the long and short run. Positive and negative shocks in energy consumption increase CO2 emissions in OECD economies in the long run. Asymmetric effects are valid for four sectors. Jel Codes: Q42, Q43, Q53