Over the past two decades, researchers have sought to establish empirical evidence for an Environmental Kuznets Curve (EKC) for carbon dioxide (CO2), with varied results. This study builds on that research to re-evaluate whether the EKC exists for CO2 emissions, using an improved dataset and the enhanced econometric technique Arellano-Bover/Blundell-Bond Generalized Methods of Moments (GMM) estimator. We directly compare OECD countries with countries of the non-OECD regions of Latin America, Asia, and Africa to determine how various factors like economic growth, population, trade, urbanization, and energy use influence CO2 emissions. We find that the OECD countries have an N-shaped curve with income growth whereas the regions of Asia and Africa experience an income-based EKC pattern. The results further reveal that population growth has a mixed impact on CO2emissions,increased trade and urbanization contribute to CO2 emissions for most areas, and increased energy use actually helps to decrease CO2 emissions.