Accounting for Population in an EKC for Water Pollution

Abstract

This paper explores the role of population in empirical studies. While panel estimation should control for differences across countries as more populated countries should pollute more, more pollution may not lead to more pollution per capita. Two models are estimated each of which control for population in different ways. One model accounts for population by expressing BOD in per capita terms and the second model regresses BOD on population using the resid- ual in the EKC regression. The Akaike Information Criterion (AIC) reveals the second model is preferred to the first model. In addition, although both models reveal an EKC, the turning points are vastly different. Future EKC studies can benefit by testing EKC models that control for population in different ways.