LAUSR

Excessive price fluctuations would affect the effectiveness of Emission Trading Scheme (ETS) and low-carbon investment. Therefore, the drivers of carbon prices need to be disentangled to analyze the price formation process, which is important for both policy makers and investors. By applying the Ensemble Empirical Mode Decomposition (EEMD) method, we decompose the historical carbon price data of the five ETS pilots in China into five groups of the independent Intrinsic Mode Function (IMF) sequences and the residue, respectively. Then, the IMFs and the residue in each pilot are reconstructed into a high frequency component, a low frequency component and a trend component, thus disentangling the effects of short-term market fluctuations, significant events, and the long-term trend. The main findings are as follows. First, the IMF with a period around one year is the most influential factor, which reflects that pilots are characterized by the yearly cycle. Second, significant events have greater impacts than short-term market fluctuations, and are the dominant driver in Shanghai and Beijing pilots. Third, the long-term trend plays a decisive role in Shenzhen, Guangdong and Hubei pilots. The price stabilization mechanism is critical to avoid a severe imbalance between demand and supply in the long run.