The power industry is the largest CO 2 emitter within the G20, accounting for over 40% of the group's total in 2023. As the G20 produces more than 80% of global emissions, assessing the effectiveness of Emissions Trading Systems (ETS) in this sector is crucial. This study examines the causal impact of ETS adoption on power-sector CO 2 emissions in G20 economies from 1990 to 2023 using multi-period staggered difference-in-differences (DID) models. ETS is modeled as a policy treatment variable, with macroeconomic controls and Environmental Kuznets Curve (EKC) terms included. Robustness is verified through placebo tests, heterogeneity checks, and dynamic group-time average treatment effects (ATTs). Results show that ETS implementation significantly reduces power-sector emissions, with effects robust across model specifications and sub-samples. More importantly, our cross-national comparative analysis reveals divergent outcomes: while dynamic analysis confirms that ETS effectiveness strengthens over time in most economies, this trend is notably absent in key exceptions such as China and Mexico, where emissions continue to rise. Overall, while ETS is an effective market-based instrument, its success is highly conditional on national contexts and structural factors. • Staggered DID confirms ETS reduces G20 power emissions, with effects strengthening over time. • Outcomes diverge: emissions fall in developed nations but rise in coal-dependent emerging economies. • Coal lock-in and intensity-based caps dampen ETS effectiveness in rapidly industrializing contexts. • Integrating the EKC hypothesis into the causal framework validates robust policy impacts.
Xie et al. (2026) studied this question.