Genişletilmiş Çevresel Kuznets Eğrisi Modelinde İçsellik Problemi: Panel Kantil Araç Değişken Yaklaşımı

Abstract

In this study, the validity of the environmental Kuznets curve hypothesis was investigated with the approach of panel quantile regression with non-additive fixed effects using instrumental variables for different country groups (Annex II, Non-Annex I, Total) defined in the United Nations Framework Convention on Climate Change for the period of 1995-2010. An augmented environmental Kuznets curve model was estimated and, in this model, CO2 emission per capita was determined as an indicator of environmental pollution and besides economic development as well as the trade volume, the share of the service sector, fossil fuel consumption and urbanization variables were taken as explanatory variables. The endogeneity problem that may arise due to possible bilateral relationships between CO2 emission-economic development and CO2 emission-trade and the heteroscedasticity problem because of the structure of the panel data are also taken into consideration with this approach. Moreover, the heterogeneous effects of explanatory variables on CO2 emissions were observed in the country groups as well as within-country groups. Although the findings obtained from the study are consistent with theoretical expectations, the environmental Kuznets curve hypothesis is not valid for any country group.