The Role of China’s ETS in Power Sector Decarbonisation

About this report

The People’s Republic of China (“China”) officially launched its national emissions trading system (ETS) in 2017, and it will come into operation in 2021. Initially covering the power sector, which accounts for over 40% of China’s energy-related CO2 emissions, the ETS is set to subsequently be expanded to other energy-intensive sectors. China’s national ETS could be an important market-based instrument to help the country meet its recently enhanced climate goals to have CO2 emissions peak before 2030 and achieve carbon neutrality before 2060.

This report explores how China’s ETS can spur emissions reductions from electricity generation and support power sector transformation. It builds on understanding of power sector development and policy trends and relies on in-depth national and provincial scenario modelling of China’s power system from 2020 to 2035. This study also analyses how the ETS’s output- and rate-based design affects overall power sector emissions, technologies and costs, and regional distribution. Finally, it recommends ways China’s ETS can play a stronger role in incentivising cost-effective and structural power sector decarbonisation to support the country’s long-term climate ambitions.
This is an extract, full report available as PDF download
Executive summary

China recently made major announcements concerning its more ambitious medium- and long-term climate goals. At the United Nations General Assembly in September 2020, President Xi’s declaration of the People’s Republic of China’s (“China”) aims to have CO2 emissions peak before 2030 and achieve carbon neutrality before 2060 set a ground-breaking vision for the country for the next four decades. China also announced in December 2020 that it would enhance its Nationally Determined Contribution (NDC) under the Paris Agreement for 2030, including reducing its CO2 emissions intensity per unit of GDP by more than 65% from the 2005 level, increasing the share of non-fossil fuels in primary energy consumption to around 25% and expanding the total installed capacity of wind and solar power to over 1 200 GW. The 14th Five-Year Plan (FYP) stipulates formulation of an action plan to peak CO2 emissions before 2030 and adoption of stronger policy measures in an effort to reach carbon neutrality before 2060.

In this context, China's emissions trading system (ETS) can be an important market-based tool to help the country achieve its climate goals and energy transition. China’s national ETS was officially launched in 2017 and will come into operation in 2021 in the power sector, before being expanded to cover other energy-intensive sectors. Even in its initial phase, it will be the world’s largest ETS, covering coal- and gas-fired power plants that are responsible for over 40% of China’s CO2 emissions from fossil fuel combustion.

China’s ETS currently employs output- and rate-based allowance allocation,1 whereas mass-based ETSs, such as the EU-ETS and California’s Cap-and-Trade Program, have a predetermined absolute cap on emissions levels covered. Allowances in China’s ETS are allocated based on a unit’s actual generation during the compliance period (e.g. total MWh of electricity generated in 2019‑2020) and predetermined emissions intensity benchmarks for each fuel and technology (e.g. CO2 emissions per MWh set for each type of coal- and gas-fired power plant). Allowances are currently allocated for free, with the introduction of auctions a future possibility (MEE, 2021). At the end of 2020, the Ministry of Ecology and Environment (MEE) released the allowance allocation plan for the power sector, with the first compliance obligations covering 2019 and 2020 emissions (MEE, 2020a).

This report explores how China’s ETS can spur emissions reductions from electricity generation and support power sector transformation. It builds on understanding of power sector development and policy trends and relies on in-depth national and provincial scenario modelling of China’s power system from 2020 to 2035.

Analysis is based on a capacity expansion and dispatch model that minimises total power system costs2 under technical, resource and policy constraints. The model assumes economic dispatch for China’s power system from 2025 onwards and expanded interprovincial trade. For wind and solar PV, feed-in tariffs (FITs) for newly installed capacity are assumed to be phased out after 2020, but new policies are assumed to be implemented to support continuous capacity expansion.

The model implements the ETS with an output- and rate-based allocation design, with the number of allowances calculated according to electricity generation and technology-specific benchmarks for four categories of coal- and gas-fired units.3 The allowance price, which is an output of the model, reflects the marginal cost of emissions abatement that minimises total system costs to meet the allocated number of allowances. The allowance price depends strongly on the stringency of the benchmarks.

This study analyses three scenarios to evaluate potential ETS impacts on China’s power sector.

  • The No-Carbon-Pricing Scenario is the counterfactual scenario against which the role of the ETS is evaluated.4 The No-Carbon-Pricing Scenario incorporates no specific policies to control CO2 emissions (i.e. neither an ETS nor command-and-control policies such as emissions caps or energy consumption standards), but it assumes economic dispatch from 2025 and policy support for wind and solar PV capacity deployment.
  • The ETS Scenario is the main scenario for assessing the role of China’s ETS in the power sector. In addition to the assumptions in the No-Carbon-Pricing Scenario, the ETS Scenario implements a national ETS with free, output-based allowance allocation for electricity generation from 2020 onwards. It also assumes that benchmarks for all coal-fired technologies become more stringent over time. As in China’s current ETS allowance allocation plan, gas-fired units with an allowance deficit are not required to purchase allowances for compliance.
  • The ETS Auctioning Scenario explores the potential effects of gradually introducing emissions allowance auctioning into the ETS. This scenario adopts the same output-based allowance allocation mechanism and benchmark-tightening trajectory as the ETS Scenario. Auctions are assumed to be introduced in 2025, moderately reducing the share of freely allocated allowances in the system first by 10%; then by 30% in 2030; and by 50% in 2035.
Key findings

With benchmarks that are gradually tightened (i.e. lowered), China’s national ETS can have an important role in reversing the upward trend of CO2 emissions from electricity generation, supporting power sector emissions to peak well before 2030. This would be essential to achieve China’s goal of attaining economy-wide peak CO2 emissions before 2030 and would contribute to the country’s ambition to reach carbon neutrality before 2060. 

CO2 emissions from electricity generation by scenario, 2020-2035


Under increasingly stringent benchmarks in the ETS Scenario, the allowance price would rise gradually from around CNY 100/t CO2 (USD 15/t CO2) in 2020 to CNY 360/t CO2 (USD 52/t CO2) by 2035. China’s annual CO2 emissions from electricity generation in 2035 would be 12% lower under the ETS than in the No-Carbon-Pricing Scenario (a drop of around 570 Mt CO2, equivalent to Canada’s total CO2 emissions from fuel combustion in 2018).

The ETS would drive emissions reductions mainly by improving the efficiency of coal-fired power generation, particularly between 2020 and 2030, and by enlarging the deployment of carbon capture, utilisation and storage (CCUS) in the power sector from 2030. With technology-specific benchmarks and free allocation, the impact of the ETS on fuel-switching away from coal is nevertheless limited.

Factors yielding additional emissions reductions in the Emissions Trading System Scenario compared with the No-Carbon-Pricing Scenario, 2025-2035


In combination with the power sector reform, an ETS with free allowance allocation could achieve these emissions reductions by 2035 while the average electricity cost remains at the 2020 level.5 In addition, the ETS would be more cost-effective than using mandatory energy consumption standards for coal-fired power plants, delivering the same emissions reductions at lower additional system costs.

With its installed capacity having increased fourfold since 2000 to reach 1 007 GW in 2018, China’s coal-fired power fleet is today the world’s largest, as well as one of the youngest and most efficient (IEA, 2020a). Nevertheless, less-efficient units such as subcritical units still represent almost half of China’s operational coal-fired power capacity. Managing its coal-fired fleet will be essential for China to achieve its emissions reduction objectives and clean energy transition.

With its output-based allowance allocation design, the ETS will prompt more efficient unabated coal-fired power generation, as units achieving emissions intensities below the applicable benchmark could sell surplus allowances while those exceeding the benchmark would need to purchase them. The benchmarking approach and the shift to economic dispatch would encourage high-efficiency units to run significantly more than they currently do. In the ETS Scenario, generation from ultra-supercritical units accounts for 66% of the coal-fired power mix in 2025 and 94% of unabated coal-based generation by 2035. Meanwhile, less-efficient and (usually) older units would either serve as back-up capacity with low annual running hours or be retired.

In addition to changing operating patterns, the ETS can accelerate the replacement of less-efficient units by the most high-performing ones. In the ETS Scenario, nearly 150 GW of subcritical, high-pressure and circulating fluidised bed (CFB) units would retire between 2020 and 2030, 43% more than in the No-Carbon-Pricing Scenario. The drive for efficiency under the ETS would increase the capacity factor of more-efficient coal-fired units, but it could also incentivise more construction of new efficient coal-fired power capacity than the No-Carbon-Pricing Scenario by 2030.

The average energy consumption of unabated coal-fired units could fall to 275 grammes of standard coal equivalent per kWh (gce/kWh) by 2035 in the ETS Scenario, which would be an 11% reduction from the 13th FYP target of 310 gce/kWh for coal-fired units in operation in 2020. As a result, the emissions intensity of unabated coal-fired power generation could decrease to 764 g CO2/kWh, 5% below the projected level without ETS implementation.

The current ETS allowance allocation design has the potential to promote CCUS technology deployment in the power sector from 2030 onwards by allowing units equipped with carbon capture technology to gain revenues by selling surplus allowances. If applying the benchmark for large coal-fired power units to CCUS-equipped units, the ETS could provide a substantial financial incentive for coal+CCUS technology and make it cost-competitive in certain regions by 2030.

In the ETS Scenario, generation from CCUS-equipped coal-fired units would account for 3% of total coal-fired power generation by 2030 and 8% by 2035. By displacing more than 470 TWh of unabated coal-fired power generation, the deployment of this technology could avoid nearly 300 Mt CO2 of electricity generation emissions in 2035 and reduce the average emissions intensity of coal-fired generation to around 710 g CO2/kWh.

Technical specificities of the ETS, such as allocation design and exemption rules, could drive the introduction of less-carbon-intensive technologies in co‑ordination with other support policies. 

While the ETS could provide an incentive for China’s coal-fired power fleet to become more efficient and potentially use CCUS technology, its promotion of gas-fired power and non-fossil alternatives would be limited under the current output-based design with multiple benchmarks and free allocation.

The output-based design grants units allowances in proportion to their production activities, encouraging plants in each technology category to reduce their emissions intensity to below the applicable benchmark so that they gain an allowance surplus rather than a deficit. Under a certain allowance price, the effective carbon cost (e.g. in CNY per kWh of generation produced) applying to a power unit would also depend on its performance relative to the benchmark. While having multiple benchmarks could help address distributional effects among technologies, they actually differentiate the effects of emissions trading for the various technologies even more than using a single benchmark would (Goulder et al., 2020).

As the benchmarks for coal- and gas-fired units are separate and China’s ETS currently does not include non-fossil sources directly, entities covered by the ETS could receive surplus allowances for coal-fired units with relatively low emissions intensities but would not necessarily gain any revenue by switching from coal to gas or nuclear or renewables. Meanwhile, under an output-based design with free allocation, units need to pay for allowances only if they perform below the applicable benchmark and have an allowance deficit, which also limits the effective carbon cost imposed on emitting units and thus reduces the incentive for fuel switching.

In the ETS Scenario with free output-based allocation and technology-specific benchmarks, generation from gas and non-fossil sources would be only marginally higher than in the No-Carbon-Pricing Scenario by 2035. As wind- and solar-based generation would not be specifically encouraged, they would remain at roughly the same level in 2035 in both scenarios. Capitalising on the ETS’s untapped potential to encourage fuel switching could further enhance its ability to drive emissions reductions and power sector transformation. 

Under an output-based ETS, total emissions are not limited by a fixed cap but depend on production activities and benchmarks applied. With free allowance allocation, the carbon cost imposed by the ETS remains limited, as only entities facing allowance deficits need to purchase allowances for compliance. Conversely, introducing auctions would require most entities to purchase a certain amount of allowances, raising the effective carbon cost faced by emitters and making it less attractive to raise the allowance volume through production choices. Auctioning could thus reduce emissions even further.

In the ETS Auctioning Scenario, with the share of auctioning increasing after 2025 and gradually reaching 50% in 2035, carbon emissions from electricity generation could peak at a lower level than under free allowance allocation, reducing annual electricity generation emissions by an additional 10% (nearly 500 Mt CO2) in 2035. Auctioning could therefore cut electricity system CO2 emissions to below the 2020 level by 2035.

Implementing allowance auctions would strengthen the competitiveness of renewables-, nuclear- and gas-based technologies vis-à-vis coal-fired plants, leading to faster decommissioning of existing coal-fired units and fewer installations of new ones. Moderate auctioning could reduce the share of unabated coal-fired power generation in the mix to below 40% by 2035, compared with nearly 50% in the ETS Scenario. Compared with free allowance allocation, auctioning would double gas-fired generation in 2035 and increase generation from wind (by 10%) and solar (by over 40%). The higher the share of allowance auctioning, the deeper and quicker power sector decarbonisation is likely to be. 

Factors yielding additional emissions reductions in the Emissions Trading System Auctioning Scenario compared with the No-Carbon-Pricing Scenario, 2025-2035


Moderately raising the share of allowances auctioned over time would keep total system cost increases in check while creating revenues that could be used towards the clean energy transition and technology development as well as to address electricity affordability and equity. In the ETS Auctioning Scenario, annual revenues generated by allowance auctioning could reach CNY 685 billion (USD 99 billion) in 2035, counterbalancing a substantial portion of the increase in total system costs.

Equity concerns at the regional level could emerge if allowance surpluses and deficits are distributed unevenly, depending on a region’s generation mix.

In the ETS Scenario, regions with a higher share of ultra-supercritical units could benefit from the ETS in 2020, while those with more subcritical and high-pressure units were likely to face additional costs. With the tightening of benchmarks and greater CCUS deployment, regional distributional effects could change significantly and widen over time. Regions with CCUS-equipped coal-fired units have the potential to gain high allowance surpluses while all unabated coal-fired power technologies would accrue deficits by 2035. Combined with a higher allowance price, the monetised impact could further widen regional disparities. Addressing potential equity issues could be important to strengthen fairness and political acceptance of the ETS.

Policy recommendations

The national ETS becoming operational is an important step in China’s climate policy development and expanded market mechanism use. To help the ETS incentivise more cost-effective and structural power sector decarbonisation, and to further align its short- and medium-term effects with China’s ambitions of reaching peak emissions before 2030 and attaining carbon neutrality before 2060, China could:

  1. Tighten its ETS benchmarks and gradually merge them to enhance the effectiveness of the output-based design.
  2. Accelerate power market reform to amplify the effects of the ETS.
  3. Introduce allowance auctioning to provide stronger signals for fuel switching and to generate revenues.
  4. Transition to a mass-based design with a fixed cap to guarantee emissions trajectory certainty and support its emissions-peaking and carbon-neutrality goals.
  5. Strengthen policy co‑ordination for ETS implementation in the power sector and its expansion to other industrial sectors, e.g. co‑ordinate it with renewables deployment, energy efficiency and CCUS support policies.

Stringent benchmarks will be essential for an output-based ETS to drive power sector decarbonisation, as benchmarks guide the emissions intensity trajectory and determine the total number of allowances for given output levels.

Gradually lowering the benchmark values will be crucial for the ETS to be consistently effective and to support China in meeting its climate goals. As average fleet efficiency improves as older units are retired and a greater share of generation comes from more efficient technologies, the average emissions intensity of thermal power generation will decrease over time. The benchmark trajectory should integrate such improvements and reduce the risk of over-allocation while continuing to provide further incentives to achieve the intended transition objectives. Depending on the stringency of the initial benchmarks and the evolution of the power fleet, the rate of benchmark tightening could be gradually adjusted for a smooth transition, before being accelerated to meet higher policy ambitions.

In parallel, merging benchmarks will reduce the differentiation of carbon price signals for different technologies and guide more cost-effective emissions reductions. Gradually transforming the multi-benchmark design into a single benchmark will optimise emissions abatement options across a larger group of technologies and assets, increasing the economic efficiency of the ETS and its encouragement of fuel switching. This would reduce the risk of providing financial incentives to high-emitting assets whose emissions could be locked in over the long term.

Having a clear benchmark trajectory will provide visibility and certainty for market participants, guide plant management and investment decisions (including for technology innovation) and accelerate power sector decarbonisation. 

Having the similar goal of promoting the use of efficient, low-emissions and least-cost resources, ongoing power market reforms and the ETS should be co‑ordinated to be mutually supportive.

Reforms favouring least-cost dispatch will be particularly vital for the ETS to be effective. Power dispatch and pricing in China continue to be determined largely by administrative mechanisms, but wide-ranging reforms aim to enlarge the role of market-based mechanisms and important progress has been made, including in pilot spot markets in several regions. Accelerating the dispatch reform would better enable the ETS and amplify its value, as a merit order dispatch system would take account of the carbon cost imposed by the ETS on less-efficient units and authorise lower-emitting technologies to operate more often. If reform progress is slow, however, and electricity generators are unable to adjust their operations based on ETS price signals, the ETS’s effectiveness in curbing emissions could be considerably constrained.

Meanwhile, the ETS can support power market reforms by integrating carbon costs into dispatch decisions and providing incentives for plants to operate more flexibly depending on their CO2 emissions levels. If externality costs (such as that of carbon) are not taken into consideration and high-emitting sources remain cost-competitive, the power market reform might optimise the cost of electricity production in a manner not necessarily aligned with the transition to a low-carbon electricity mix. 

Allowance auctioning in China’s output-based ETS would make it more attractive to switch to non-fossil and gas fuel sources, driving more fuel switching and amplifying the impact of the ETS. As auctioning raises the carbon cost imposed on CO2-emitting technologies, it affects production from high-emitting sources and therefore reduces total associated emissions. By making gas-fired and non-fossil technologies more competitive, allowance auctioning would make the ETS more effective in driving power sector transformation.

Gradually phasing in allowance auctioning according to a clear timeline would accelerate the energy transition while allowing market participants time to adapt to the system and could keep electricity cost increases moderate. Part of the revenues generated by the auctions could be used to address electricity affordability and the distributional effects of the ETS, and they could also be invested in low-carbon technology development to foster more rapid decarbonisation. 

As China aims to have its emissions peak before 2030 and to reach carbon neutrality before 2060, limiting its total emissions in addition to reducing emissions intensity will be essential. When it expands the ETS beyond power to other industrial sectors, using the current output-based allowance allocation design would likely be more complicated and challenging than adopting a fixed-cap system.

Transitioning to a mass-based design with an absolute cap would provide significantly more certainty for controlling emissions from sectors covered by the ETS, reduce the risk of encouraging the construction of additional high-carbon assets, and ensure coherence with China’s economy-wide emissions-peaking and carbon neutrality plans. Using a mass-based design would also allow the ETS to send uniform carbon price signals and promote the most cost-effective choices for emissions reductions (Goulder et al., 2020), including by providing further incentives for switching to low-carbon energy sources. The initial phase of ETS implementation will improve emissions monitoring and supply valuable information for setting an absolute ETS cap and mapping out its trajectory.

To achieve the low-carbon energy transformation required for carbon neutrality before 2060, China will need to implement numerous ambitious policies aimed at a variety of objectives.

Figuring within a complex policy landscape, the ETS interacts with various economy-wide and sector-specific mechanisms. Strengthening co‑ordination between the ETS and other policy instruments, such as those affecting renewables deployment, energy efficiency and technology innovation (e.g. CCUS), could increase the effectiveness of many policies and achieve a more cost-effective and impactful outcome, whereas a lack of co‑ordination might lead to duplicated or counterproductive policy efforts (IEA, 2020a).

Co‑ordinating the ETS with other market-based policies, such as the Chinese Certified Emissions Reduction (CCER) offsetting scheme, renewable portfolio standards or green power trading certificates, could enhance mitigation and accelerate the low-carbon energy transition while reducing the overall costs of the transition.

The ETS, through its allocation design and by using market forces, could provide influential price signals to accelerate the deployment of innovative and low-carbon technologies. This analysis demonstrates the potential of an ETS to spur CCUS deployment, but it could also encourage the use of other new low-carbon technologies such as emerging renewables and utility-scale storage with adapted design.

However, ETS price signals would be effective only if accompanied by policies to reduce investment risks for these nascent technologies. In the case of CCUS, an ETS with output-based allocation could provide an important financial incentive, but near-term companion policies would be needed to: create favourable investment conditions by offering direct support for early CCUS projects; co-ordinate the development of CCUS hubs and incentivise the construction of CO2 storage in key regions; and support research and development (R&D) and demonstration projects to further improve CCUS performance and reduce costs.

In addition to supporting power sector decarbonisation, an ETS can serve as an umbrella policy to enable cost-effective emissions reductions in emission-intensive sectors. After a short period of initial application to the power sector, national ETS coverage should be rapidly expanded to key industrial sectors. The ETS could thus provide pricing incentives for energy efficiency measures and encourage demand-side switching to low-carbon sources. Exploring synergies with energy efficiency mechanisms such as energy efficiency obligation schemes could raise the effectiveness of both policies.

With expanded emissions coverage and better co‑ordination with other energy and climate policies, the ETS could become a primary policy instrument to deliver cost-effective emissions reductions, accelerate emissions-peaking and foster energy sector decarbonisation to realise China’s long-term goal of carbon neutrality.

  1. A rate-based ETS is often termed a tradable performance standard.

  2. Total system costs include annualised capital costs and operating costs for electricity generation, as well as costs of balancing electricity supply and demand and of transmission.

  3. Benchmark trajectory design is explained in Chapter 2. Assumptions for 2020 benchmark values differ from the official benchmarks, as they were defined before China’s ETS 2019-2020 allowance allocation plan for the power sector was published. Analysis in this report is based on scenario modelling, and findings on structural implications of the ETS remain relevant for other initial benchmark values.

  4. The No-Carbon-Pricing Scenario serves as the baseline scenario to assess the ETS’s effects and potential. It differs from the World Energy Outlook (WEO) Stated Policies Scenario (STEPS), which reflects the impact of existing policy frameworks and announced policy intentions and includes carbon prices for China’s power, industry and aviation sectors. 

  5. The electricity cost reflects the average system cost per unit of electricity generated. Under free allocation, entities receive allowances for free, and total allowance surpluses and deficits among entities balance out at the system level, limiting the increase in system costs.