Has the European carbon market really reduced emissions from the electricity sector?
At a time when the World Bank is taking stock of carbon pricing in 2025, a retrospective study of the electricity sector over the three regulatory periods of the European Union Emissions Trading System provides insightful findings. Although it had no net effect at the outset, it subsequently led to significant reductions in emissions.
01/07/2026
4 min
Share
Since 2005, the European Union Emissions Trading System (EU ETS) has required emitters to pay for emissions via allowances to encourage their reduction. This system represents the first major multi-country carbon market and a model that is now being observed around the world.
According to the World Bank’s ‘State and Trends of Carbon Pricing 2025’ report around 28% of global emissions are covered by a direct carbon price, with 80 instruments in place (taxes and carbon markets/trading systems), generating more than $100 billion in annual public revenue over the past two years. Globally, the electricity sector is the sector with the highest carbon pricing coverage: more than half of its emissions are already subject to a carbon price.
The report also highlights a number of current trends. These include the expansion of existing systems, but also the rise of emissions trading schemes (ETS) in major emerging economies (Brazil, India, Turkey). In Europe, the creation of an ‘ETS 2’ will cover fuels, buildings and road transport from 2027.
This raises a simple but crucial question for policymakers: what does a rigorous ex-post analysis of the EU ETS tell us about its ability to reduce emissions, at least where it matters most: electricity production?
Our study, recently published in the journal Ecological Economics, assesses the effectiveness of the EU ETS on emissions from the electricity sector during its first three completed regulatory periods (2005–2020), across 24 Member States.
The methodological challenge is simple to express but difficult to resolve, as we only have access to historical emissions data with the ETS. To properly measure its impact, we would need data showing what the level of emissions would have been without such a system in place. However, in the electricity sector, many factors change at the same time: overlapping policies (support for renewable energies, standards, etc.), weather, demand, energy prices, etc.
To avoid comparison with an arbitrary ‘control group’, we constructed a credible reference scenario. In practice, this is an internally derived counterfactual of the European electricity sector emissions, based solely on observable factors that are not influenced by the ETS. For example: temperature, demand, wind/solar production, international oil, gas and coal price indexes.
We then compared historical emissions with those predicted by this scenario. The difference between the two-time series reveals the effect of the ETS month by month and also phase by phase.
Three phases, three results
We observe distinct effects across the three phases studied.
- The first (pilot) phase, from 2005 to 2007, shows no statistically significant impact on emissions from the electricity sector. This result can be explained in particular by an overly generous supply of (emissions) allowances at the outset, which weakened the price signal.
- In the second phase, from 2008 to 2012, we observe an average reduction of around 12% in emissions compared to the scenario without ETS.
- Finally, in the third phase (2013-2020), the reduction was around 19%. This increased efficiency coincided with ETS design reforms, in particular abandoning free allowances for electricity generation and the tightening of the cap.
By aggregating the period 2005–2020 as a single ‘policy shock’, we also found a significant reduction (~15%) consistent with the phase analysis.
These results take economic cycles into account, and control for the influence of other factors (such as overlapping national policies). Therefore, they highlight that, once these elements are taken into consideration, the carbon market delivered its intended effects, and increasingly so as it has become more demanding.
The electricity sector, which is a major source of greenhouse gas emissions in many regions, is subject to numerous (overlapping) policies that complicate the assessment of the carbon market’s effectiveness. This is particularly true as the decarbonization of the electricity sector multiplies downstream effects (industry, mobility) as the economy becomes more electrified – a trend highlighted by the World Bank. This is precisely why a robust methodology is so important.
Lessons for other carbon markets
These lessons have implications far beyond Europe: as major economies establish or strengthen their carbon markets, the credibility of the price signal and the quality of the framework (a truly binding cap, mechanisms to avoid over-allocation, auctions) appear to be essential conditions for success.
Our phase-by-phase analysis also shows that carefully targeted reforms can transform a system that is initially too broad into an effective tool. As new markets emerge and interconnect thanks to policies such as carbon border adjustments, these benchmarks can guide regulatory choices.
In this context of rapid expansion of carbon pricing – which provides massive funding for the green transition – such assessments remain essential to anchor the next steps in evidence rather than intentions.
The European Union is indeed preparing to extend the ETS system to fuel combustion from building and road transport from 2027, using an ‘upstream’ approach – i.e. placing the obligation on fuel suppliers, before final consumption – already applied by certain carbon trading systems (California, New Zealand).
Our study complements work carried out in other sectors and provides a useful empirical basis. Ex post, sector-by-sector evaluation of what works and why remains the best guarantee of an effective and socially acceptable expansion of carbon markets.
This article by Professor Ethan Eslahi is the English translation of an article originally published on the Conversation France.
