Carbon leakage in emissions trading systems: a systematic review and meta-analysis of ex-ante and ex-post evidence

Emissions Trading Systems (ETSs) are a widely adopted policy tool for reducing greenhouse gas emissions. However, emission reductions achieved within ETSs can be partially offset by increases in emissions elsewhere , a phenomenon known as carbon leakage. This paper systematically reviews the global literature on carbon leakage associated with ETSs through a combined quantitative meta-analysis and qualitative synthesis. Findings show that ex-ante modelling studies generally report positive carbon leakage rates, with estimates varying widely across schemes. Leakage rates are typically lower when more of the global emissions are covered, anti-leakage measures are included, and when using a computable general equilibrium (CGE) model. Ex-post studies find leakage through channels such as interstate electricity trade in the U.S. and investment relocation in the EU ETS and China pilot ETSs. Findings on goods trade and carbon transfer within firms are mixed. These findings highlight the differences in results between ex-ante and ex-post studies, the heterogeneity of leakage outcomes across schemes, and the critical role of country-specific context in designing ETSs and associated anti-leakage measures. Key policy insightsThe systematic review finds that ex-ante studies estimate a lower leakage rate when emissions coverage is broader, anti-leakage measures are included, and computable general equilibrium models are used.Ex-post studies evidence leakage through specific channels including investment leakage in the EU ETS and Chinese pilot ETSs, and coal-to-gas substitution in the U.S. regional ETSs.Jurisdictions designing a mitigation policy should account for context-specific leakage risks when evaluating alternative policy instruments. Further ex-ante evidence on leakage is needed for emerging ETSs beyond the major schemes.To comprehensively address carbon leakage, a broader, integrated policy mix incorporating both demand- and supply- side anti-leakage measures, enhanced international cooperation, and sector-specific supports is recommended. The systematic review finds that ex-ante studies estimate a lower leakage rate when emissions coverage is broader, anti-leakage measures are included, and computable general equilibrium models are used. Ex-post studies evidence leakage through specific channels including investment leakage in the EU ETS and Chinese pilot ETSs, and coal-to-gas substitution in the U.S. regional ETSs. Jurisdictions designing a mitigation policy should account for context-specific leakage risks when evaluating alternative policy instruments. Further ex-ante evidence on leakage is needed for emerging ETSs beyond the major schemes. To comprehensively address carbon leakage, a broader, integrated policy mix incorporating both demand- and supply- side anti-leakage measures, enhanced international cooperation, and sector-specific supports is recommended.