Revisiting environmental kuznets curve for carbon dioxide emissions: The role of trade.

• Re-examination of the EKC hypothesis, taking into account the role of international trade. • International trade postpones the peak time of global CO 2 emissions. • The validity of the EKC for global CO 2 emissions, the role of trade cannot be overstated. • Based on the production principle and on the consumption principle, 39 countries can be classified into 5 groups. In the three decades since the environmental Kuznets curve (EKC) hypothesis was proposed, there have been a massive amount of empirical studies on the relationship between per capita income and carbon dioxide (CO 2) emissions based on the production accounting approach. International trade has played a critical role in effectively and efficiently allocating resources in the process of economic globalization. Through international trade, part of the responsibilities for CO 2 emission mitigation which should have been taken by consumers have been transferred to the producers of traded goods and resources. The focus of this research is on the re-examination of the EKC hypothesis, taking the role of international trade into account. The findings are as follows. In the first stage, we calculated the CO 2 emissions of 39 countries using the input-output analysis approach based on production accounting and consumption accounting. Currently, China is the largest CO 2 emitter, surpassing the U.S. However, the per capita CO 2 emissions of China lag behind those of the U.S. In the second stage, although both an inverted U-shaped curve and an N-shaped curve were obtained, the cubic functional form model is better fitted. In addition, the turning point occurs earlier if adopting the production-based accounting rather than adopting the consumption-based accounting, implying that international trade postpones the peak time of global CO 2 emissions. From the above analysis, it follows that when testing for the validity of the EKC for global CO 2 emissions, the role of trade cannot be overstated. [ABSTRACT FROM AUTHOR]