Simultaneity Modeling Analysis of the Environmental Kuznets Curve Hypothesis

The environmental Kuznets curve (EKC) hypothesis has been recognized in the environmental economics literature since the 1990's. Various statistical tests have been used on time series, cross section and panel data related to single and groups of countries to validate this hypothesis. In the literature, the validation has always been conducted by using a single equation. However, since both the environment and income variables are endogenous, the estimation of a single equation model when simultaneity exists produces inconsistent and biased estimates. Therefore, we formulate simultaneous two-equation models to investigate the EKC hypothesis for fifty-six countries, using annual panel data from 1990 to 2012, with the end year is determined by data availability for the panel. To make the panel data analysis more homogeneous, we investigate this issue for a three income-based panels (namely, high-, middle-, and low-income panels) given several explanatory variables. Our results indicate that there exists a bidirectional causality between economic growth and pollution emissions in the overall panels. We also find that the relationship is nonlinear and has an inverted U-shape for all the considered panels. Policy implications are provided.