欧盟碳排放权市场重大公告事件对碳价格的影响

Impact of Pivotal Announcement Event on Carbon Price in European Union Emission Trading Scheme

随着全球气候变化问题不断突出,建立碳排放权交易市场成为各国政府应对气候变化、控制温室气体排放的重要措施。作为新生市场,碳市场价格容易受到制度类重大公告事件的较大冲击。文章通过在传统AR-GARCH模型的均值方程和波动方程中加入双边修正虚拟变量,提出双边修正虚拟变量的事件研究方法;以欧盟碳市场国家配额分配计划(NAP)公告和核证排放量公告(VEA)事件为例,分析了重要公告的发布对碳价格的潜在影响。实证结果表明,双边修正虚拟变量方法可以有效捕捉公告事件对碳价收益率和波动性的事前、事后影响过程。国家配额分配计划公告对碳价收益产生了显著的正向影响,且表现出较长的事前影响。核证排放量公告在碳市场第一阶段对碳价收益产生了显著的事后负向影响,在第二和第三阶段对碳价收益产生了显著的事前和事后正向影响,且在两阶段对碳价波动性都没有显著影响。第一阶段的核证排放量信息为市场预期提供了准确参考,使第二和第三阶段的核证排放公告对碳市场的影响减弱。

With the worsening of global climate change, establishment of carbon trading scheme has become an important option for governments to address the issue of climate change. As newly created market, carbon market is easily subject to the shocks from institution modification, rule adjustment, and vital information announcement. The research establishes an event study method through modeling bilaterally modified dummy variables in AR-GARCH model to unveil the ex ante and ex post impact process and impact pattern of the announcement event. The study takes National Allocation Plans（NAP） announcement and Verified Emission Announcement（VEA） event in the European Union emission trading scheme（EU ETS） as examples. Empirical results show that bilaterally modified dummy variables can efficiently capture the ex ante and ex post impact process of announcement event on expected carbon returns and volatility. The announcement of NAP I has significant positive impact on carbon returns, and has no significant impact on carbon price volatility. The announcement of VEA I has significant negative ex post impact on carbon returns and the announcement of VEA II has significant positive ex ante and ex post impact on carbon returns. In addition, the announcements of VEA I and VEA II have no significant impact on carbon price volatility. Compared with the impact of VEA I announcement, the impact of VEA II announcement has decreased, mainly due to the fact that the verified emission of the first phase provides reliable data to calibrate expectations.