中文摘要：

作为减缓气候变化的重要调控工具之一,征收碳税在减少CO2排放的同时也将给国家经济发展带来一定负面影响。由于所采用的分析模型、数据以及情景假设等不同,在征收碳税对中国经济（GDP）影响幅度上,国内现有研究结论存在较大差异。本文利用Meta方法对导致评估结果差异的各种因素进行分析,研究结果表明模型选择和碳税税率的设定是导致评估差异的主要原因。其中,模型选择是造成研究结论差异的主要原因,可以解释结论差异的74%左右;如果考虑模型选择与碳税税率的交叉影响,其解释效力可进一步提高到87%;而征税方式与碳税使用方式对研究结论差异没有显著影响。相比于其他模型,CGE模型与MARKAL-MACRO模型的评估结果比较稳定,影响幅度居中并较为接近,更为适用于评估实施碳税对GDP的影响程度。基于研究结果,提出在分析征收碳税对中国经济影响的研究上亟需方法论创新,需要特别注重研究方法的科学性。

英文摘要：

As an effective method for mitigating the impacts of climate change, the implementation of a carbon tax not only controls and reduces carbon emissions, but can also have negative impacts on economic development. Estimation results of the implementation of a carbon tax on China＇ s GDP vary significantly among studies and far from reaching a consensus. Here, we used meta-analysis methods to analyze factors that lead to variation in estimates on China＇ s Gross Domestic Production （GDP） following a carbon tax. We classified model setting, modes of taxation, tax rate levels and usage of the revenue from carbon tax as four main factors to explain differences amongst estimates. Eighty-five samples were obtained from the literature and the results indicate that the model setting and carbon tax rate are the factors leading to different estimates. Among them, model setting is the most important reason for the differences in conclusions which accounts for about 74% of variation among studies. If the interaction effect of these two factors is further considered, the explanatory power rises to 87%. Compared to other models, results of CGE and MARKAL-MACRO models are robust and stable and the impact on GDP percentage change of these two models are similar. These two models can fully reflect the internal regulation and feedback mechanisms of the economic system, and these elements of resources such as labor and capital could be reallocated and flowed in all sectors of the economy according to the optimal rate of return. Thus, these two models are more better suited to impact evaluation.