This paper provides new empirical evidence of the effectiveness of Emissions Trading Systems (ETS) in reducing greenhouse gas emissions. Using a panel dataset spanning 1995-2023 across countries, we employ difference-in-differences framework specifically designed for staggered treatment adoption, addressing recent critiques of traditional two-way fixed effects models that can produce biased estimates when treatment effects are heterogeneous. We find that ETS implementation leads to substantial reductions in greenhouse gas emissions, with estimates suggesting a 19 percent decline in GHG emissions on average across all post-implementation periods. These effects are particularly pronounced in the power generation sector (30.2 percent reduction) and industrial combustion (27.9 percent reduction), while transportation shows more modest but still significant decreases (8.4 percent). The impact of ETS strengthens over time, with emissions declining by approximately 23.1 percent after ten years of implementation. We find heterogeneous effects across implementation cohorts. Our results remain robust when controlling for concurrent carbon taxation policies and when examining carbon intensity measures, indicating that ETS can effectively induce transitions toward less emissions-intensive production structures.

 

Funded through the 2024 CSIS DfD Lab Call for Proposals