Environmental Kuznets Curve : Panel Data Evidence from Developing Countries

This paper aims at analyzing the shape of the environmental Kuznets curve (EKC) for developing countries. We compare four different techniques of estimations, such as pooled ordinary least square (OES), fixed effect, random effect, dynamic panel data, and pool OLS at first difference. We obtain three major findings. First, state of assumptions, data span, and model specification, have important implication on sign and parameter estimate. We argue that those conditions contribute to uncertainty in parameter estimate. Second, the models share a common findings such as inverted U-curve in explaning the relationship between per capita COi emissions and per capita GDP. Second, we find that the economic transition from the agricultural to the services sector has a positive impact to ease CO2 emissions. Third, in the long run, an increase in per capita GDP will lead to much higher CO2 emissions per capita than in the short run. We suggest that due to uncertainty on turning point and most people are riskaverse, wisdom of knowledge prefer anticipatory instead of reactive policy toward global warming. A climate-friendly path needs to be a basis in economic development planning