The Environmental Kuznets Curve: some really disturbing Monte Carlo evidence

The Environmental Kuznets Curve hypothesis posits an inverse U- shaped relation between environmental pollution and income. The empirical literature has largely ignored the time series properties of the data used to test the EKC. This paper uses Monte Carlo experiments to analyse whether the order of integration influences the EKC empirical strategy. We show that if the variables used are I(1), the results will spuriously confirm the EKC hypothesis in 40% of the cases. Furthermore, accepting the EKC also influences the critical values for the rejection of no cointegration in the Engle- Granger framework.