Scholarlink

<p>To achieve the goal of carbon neutrality, technology innovation is an inevitable approach. China has introduced a pilot emission trading system (ETS) to reduce carbon emissions. However, there are limited empirical studies examining the impact of ETS on carbon-neutral technology innovation (CNTI) at city level, particularly through machine learning methods. This research aims to investigate the impact of the ETS policy pilot program on the CNTI using city-level data via both traditional econometric models and machine learning algorithms. First, random forest is employed to screen control variables. Next, both difference-in-differences (DID) method and generalized random forest (GRF) method are utilized to evaluate the causal effect of the pilot ETS policy on carbon neutral technology innovation. This is followed by analyses of driving mechanisms and heterogeneity analysis. The study finds that the pilot ETS significantly promoted CNTI, although the DID method overestimated the treatment effect compared to the GRF. Mechanism analyses reveal that the policy worked by attracting foreign direct investment (FDI) and strengthening environmental regulation. The effect is more pronounced in eastern cities and those with higher carbon prices. In view of these findings, this paper puts forward some policy recommendations.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Will carbon emissions trading promote carbon neutral technological innovation in China? An analysis combining DID and generalized random forest

  • 期刊论文

  • 发表日期：

    2026年1月24日

查看全文
  • Jingru Li,
  • Haoran Lin,
  • Yanbin Cheng,
  • Huanyu Wu

摘要

To achieve the goal of carbon neutrality, technology innovation is an inevitable approach. China has introduced a pilot emission trading system (ETS) to reduce carbon emissions. However, there are limited empirical studies examining the impact of ETS on carbon-neutral technology innovation (CNTI) at city level, particularly through machine learning methods. This research aims to investigate the impact of the ETS policy pilot program on the CNTI using city-level data via both traditional econometric models and machine learning algorithms. First, random forest is employed to screen control variables. Next, both difference-in-differences (DID) method and generalized random forest (GRF) method are utilized to evaluate the causal effect of the pilot ETS policy on carbon neutral technology innovation. This is followed by analyses of driving mechanisms and heterogeneity analysis. The study finds that the pilot ETS significantly promoted CNTI, although the DID method overestimated the treatment effect compared to the GRF. Mechanism analyses reveal that the policy worked by attracting foreign direct investment (FDI) and strengthening environmental regulation. The effect is more pronounced in eastern cities and those with higher carbon prices. In view of these findings, this paper puts forward some policy recommendations.