6.1 Overview of China’s Current Energy Structure

6.1.1 Energy Sector Produces Nearly 90% of Carbon Emissions in China; Building Green Energy Supply System is Top Priority

At the current stage, China’s energy supply is still dominated by coal, with petroleum, natural gas, and non-fossil fuels providing the rest. Reducing carbon emission and tackling climate change have become the common task of the international community, and according to UN Environment Programme’s Emissions Gap Report 2019 of the total emission, China is the largest carbon dioxide emitter and contributed over 25% of global emissions in 2018. China’s promise to achieve carbon neutrality in 2060 will facilitate its integration into the international community and help the country achieve its dual circulation model, creating synergies between domestic and overseas markets.

We believe greater efforts toward reaching “low-carbon” and “zero-emissions” are key to achieving carbon neutrality in the energy sector.

6.1.2 Total Energy Demand May Increase Under Steady Economic Growth Despite Decline in Energy Consumption Per Unit of GDP

Official data shows that China’s GDP per capita is only one-sixth to one-third of that of developed countries, pointing to substantial growth potential. CICC macro team expects that in 2020Q1 China’s GDP to grow 4.2%, 3.0%, 2.5% and 2.0% in 2030, 2040, 2050 and 2060 respectively. China’s GDP per capita rose to US$10,276 in 2019 (more than doubling from US$4,628 in 2010, see Fig. 6.1), which means it not only met twice as much the US$10,000 GDP per capita national target but has also successfully built a moderately prosperous society in 2020 in all respects (initiated by the 18th CPC National Congress). Looking ahead, given China’s plan to basically achieve modernization by 2035 and national rejuvenation by 2050, we expect its GDP per capita to rise to US$41,105 in 2060, reaching current levels in Japan and Germany.

Fig. 6.1
figure 1

Source NBS, government websites, CICC Research

Comparison of per capita GDP (China vs. other countries and regions).

China’s current energy consumption per unit of GDP is still at a high level and to be lowered by energy restructuring and conservation. According to official data, China’s energy consumption per unit of GDP stood at 0.328kg of standard coal/USD, which is higher than developed countries’ 0.116–0.264kg of standard coal/USD (Fig. 6.2). With improvements in the country’s economic structure and its greater efforts to curb energy consumption, we expect its energy consumption per GDP to drop 63% from the current level to 0.123kg of standard coal/USD in 2060, lower than current levels in the US and Japan.

Fig. 6.2
figure 2

Source National Energy Administration, NBS, government websites, CICC Research

Comparison of energy consumption per GDP (China vs. foreign countries and regions).

However, even considering the possible decline in energy consumption per capita, we expect China’s economic growth to boost total energy consumption to 6.73bnt of standard coal in 2060 (+38% compared to the current level, see Fig. 6.3). We estimate that China’s total energy consumption will likely rise to 5.76, 6.36 and 6.73bnt of standard coal in 2025, 2035 and 2060 respectively, with continuous YoY growth. As the proportion of tertiary industries with lower energy consumption per unit of GDP rising, we expect China’s energy consumption growth to be slower and its CAGR to fall from 3.3% in 2021–2025 to 1.0% in 2025–2035 and 0.2% in 2035–2060.

Fig. 6.3
figure 3

Source NBS, CICC Research

Total energy consumption forecast.

6.1.3 Carbon Emissions Reduction is Difficult to Achieve Under Current Energy Structure; The Country Needs Stronger Top-Down Planning and Policy Support

As the world’s largest emitter, China confronts multiple difficulties in lowering carbon emissions. To achieve carbon neutrality, China’s energy structure should be cleaner, safer and more economically feasible. With increasing energy demand and restrictions on carbon emissions, the goal to achieve carbon neutrality will be more challenging for China, which requires stronger policy support. It will also accelerate China’s energy transformation, benefiting economic growth eventually. Setting the goal of carbon neutrality marks China’s step toward energy reform, and will make the country’s energy structure cleaner, safer and more economically favorable in 2060. The energy structure in China is displayed in Fig. 6.4.

Fig. 6.4
figure 4

Source BP Energy, CICC Research

China’s energy structure (2019–2060).

To reach carbon neutrality in the energy industry, China should make electric power generated by non-fossil fuels the major energy source in its future zero-emission system and accelerate the rate of electrification. The government may also promote the development of new non-electric technologies such as hydrogen energy and carbon capture technology. Based on estimates by various CICC sector research teams, we expect the energy sector to achieve carbon neutrality by 2060 through raising clean electricity’s share of total energy supply to 70%, green hydrogen’s share to 8%, and the share of other energy sources supported by carbon capture technologies to 22%. Attaining this goal requires efforts from both the supply and demand sides. On the supply side, power generation should be cleaner, while on the demand side, government should encourage the deployment of other renewable technologies.

6.2 Start from More Feasible Methods to Achieve Energy Transition Through More Economical Ways

How can the energy sector contribute to carbon neutrality? Given that China should achieve green electricity first, then clean non-electric energy, raising the electrification rate should help accelerate emission cuts.

Energy transition lays the foundation for carbon neutrality. We think it will be difficult for the cement, steel and electrolytic aluminum industries to completely internalize the cost of achieving carbon neutrality given their current energy structure, technology level and costs. In contrast, the chemical industry can internalize the costs but would have to sacrifice profitability. The costs of carbon neutrality are more affordable for manufacturing industries only.

Among all the paths to achieve carbon neutrality, why do we think green electricity should come first, and then non-electric applications?

Green Electricity: We believe replacing thermal power with clean energies can effectively lower carbon emissions from power generation. We estimate the power industry’s green premium will likely turn negative by 2030, ahead of other sectors. A zero-emission electric power generation system is almost economically feasible today, and its green premiums may turn negative in 10 years with lower grid costs.

Clean Non-electric Energy: Acceleration of electrification rate, energy conservation, carbon emissions reduction, and new technology applications including hydrogen power and carbon capture, rely on technology breakthroughs and policy support. Green premiums for hydrogen power and carbon capture are likely to remain high in the near future and China should focus on electrification since it is more economically feasible. We estimate China’s electrification rate may rise to 70% by 2060, and the remaining 30% of energy demand could be met by hydrogen power and carbon capture technology. These two technologies have different applications. According to estimates by various sector teams in CICC Research, we believe the penetration rate of carbon capture will likely be higher in 2060, especially in the commodities sector. In contrast, hydrogen power will likely be applied more in the transportation and chemical sectors. Our estimation of the green premium in the non-power sector is 175% in 2021. If costs of fossil fuels stay unchanged but those of carbon capture and hydrogen power decline, green premiums in the non-power sector may fall to 57% in 2030 and 14% in 2060. While they may fall, we believe they will stay above 0%. This means that the total cost of non-electric power consumption will likely rise to achieve carbon neutrality.

As shown in Table 6.1, green electricity and clean non-electric applications combined present the best solution for carbon neutrality.

Table 6.1 Green electricity and clean non-electric applications combined present the best solution for carbon neutrality

6.3 Lowering Costs to Raise Non-fossil Fuels’ Proportion in Power Generation

6.3.1 Calculation of Green Premiums in Power Sector

Costs in the power sector mainly come from three key variables: power generation, absorption, and electricity transmission and distribution. For power generation, we look at the structure of power sources and their costs per kilowatt-hour (kWh) of electricity generation. For absorption, the proportion of flexible power resources including peak-load regulation supported by thermal power units, pumped-storage hydropower, electrochemistry energy storage, and their costs per kWh generation. Moreover, we expect the penetration rate growth of wind and photovoltaic (PV) power to boost demand for absorption, pushing up the proportion of absorption volume in power generation. Thus, compared with the base scenario, achieving carbon neutrality will change power generation costs and boost demand for absorption. We assume the costs of electricity transmission and distribution remain unchanged. We expect green premiums in China’s power industry to turn negative by 2030 (Fig. 6.5).

Fig. 6.5
figure 5

Source China Electricity Council, corporate filings, GGII, LEK, CICC Research

The green premium in the power industry in 2021 and 2023 (the CICC alternative energy and electrical equipment team expects China’s power system to achieve carbon neutrality in 2051). Note Our estimation of costs per kWh power generation is based on data from China Electricity Council, energy companies, GGII and LEK Consulting; we cross check it with electricity prices released by National Energy Administration and regional power grids.

2021: Compared to the basic scenario, in the carbon neutrality scenario, the green premium is 6% for power generation, and 3,600% for absorption. If the cost of electricity transmission and distribution remain unchanged, the green premium in the power sector would be 17%.

2030 forecast: Compared with the base scenario, power generation sees a green premium of −17% in the carbon neutrality scenario and around 5 times the size of green premiums in absorption. The green premium in the power sector may thus turn to −4% in 2030.

The CICC alternative energy and electrical equipment team expects all thermal power units to retire in 2051. By then, China’s energy source mix should include hydro, nuclear, wind and solar energies, achieving zero-emissions.

6.3.2 Power Generation: Non-fossil Energy’s Lower Costs Per kWh Fuels Transition to Cleaner Energy Mix

The costs per kWh electricity generation of hydropower and wind power are already lower than that of thermal power in 2021 (see Fig. 6.6 left side). But the nationwide average costs per kWh of nuclear power and PV power are still slightly higher than for thermal power. Thermal, hydro, wind, nuclear and PV power account for 68%, 16%, 7%, 5% and 4% of total generation, respectively. We estimate average power generation cost in China is Rmb293/kWh. In the carbon neutrality scenario, if the power system achieves zero-emissions by 2060, average green electricity costs will rise to Rmb311/kWh, with a green premium of 6%.

Fig. 6.6
figure 6

Source China Electricity Council, corporate filings, GGII, LEK, CICC Research

Breakdown of costs per kWh power generation (generation costs determine proportion). Note Our calculation of costs per kWh power generation is based on data from the China Electricity Council, energy companies, GGII and LEK Consulting; we cross check it with electricity prices released by the National Energy Administration and regional power grids.

In 2030, we estimate the costs of PV and wind power may be lower than that of hydro and nuclear power, and the cost of clean energy lower than thermal power. According to the CICC alternative energy and electrical equipment team’s power resource mix forecast (thermal, PV, wind, hydro and nuclear power contribute 54%, 14%, 13%, 12% and 7% of total power generated in 2030), we estimate China’s power generation cost would then be Rmb257/kWh in 2030. In the carbon neutrality scenario, the cost of green electricity is Rmb213/kWh under a zero-emission structure. And as the costs of wind, hydro, nuclear and solar power are lower than thermal power, the green premium of power generation would turn to −17% by 2030 (see Fig. 6.6 middle).

As the costs of clean power are lower than thermal power after 2030 and thermal power units would gradually retire, the CICC alternative energy and electrical equipment team expects emissions of the power generation system would fall without external help and would see zero emissions by 2060 (see Fig. 6.6 right side).

6.3.3 Absorption: Complementary Multi-energy System Minimizes Absorption Costs

We believe that increasing the new energy penetration rate will drive up absorption costs. Given global experiences, we notice that absorption costs (calculated by amortized ancillary power system service costs of per kWh power generation) will likely rise notably with wind power and PV power’s penetration rate growth in the power system. By then, absorption can represent 10%–20% of total costs in different power markets, hence extending the time for green premiums to fall to zero. We expect a multi-energy complementary power system would minimize absorption costs. Breakdown of per kWh absorption costs is displayed in Fig. 6.7.

Fig. 6.7
figure 7

Source Our estimation of costs per kWh power generation is based on filings, and data from GGII and LEK Consulting, we cross check it with the electricity prices released by National Energy Administration and regional power grids

Breakdown of per kWh absorption costs (costs determine structure).

As peak-load regulation supported by thermal power can minimize absorption costs, we don’t think thermal power will be abandoned in the near future.

A multi-energy complementary system effectively lowers absorption demand and costs as well as green premiums of power generation and power grid (see Fig. 6.8).

Fig. 6.8
figure 8

Source CICC Research

Wind and solar power can compensate each other; multi-energy complementary system lowers demand for energy storage.

If electrochemistry energy storage achieves technology breakthroughs and much lower prices with policy support, it may notably lower absorption costs and help China achieve green electricity earlier.

6.4 Non-power Sector: Electrification, Hydrogen Power and Carbon Capture Fuel Energy Transition

6.4.1 China’s Electrification Rate Will Likely Reach 70%; The Remaining 30% Demand Requires Non-electric Power and Other Energy Resources Supported by New Technologies to Achieve Carbon Neutrality

Looking into sub-sectors, black metal (iron and steel), transportation and warehousing (AFV, railway) and civil sectors (collective heating) have more room for growth in electrification (Fig. 6.9). Based on our estimation, non-electric power consumption accounted for 54% of China’s energy consumption in 2019. However, some energy cannot be replaced by electric power given higher requirements for energy density and long-term storage, and the fact that it cannot transform to electric power. Based on the estimation from the CICC research team of the future form of energy, together we see the electrification rate in 2060 may reach 70%, with other energy sources meeting the remaining 30% demand.

Fig. 6.9
figure 9

Source National Energy Administration, CICC Research

Electrification rate forecast for major energy consumption sectors (2017 vs. 2060).

6.4.2 Non-power Sector: Hydrogen Power and Carbon Capture is Feasible Technology Solution to Carbon Neutrality

Achieving carbon neutrality in the non-power sector is the last step towards overall carbon neutrality. We think carbon capture and hydrogen power are two feasible technologies at the current stage. Carbon capture’s penetration rate will likely be higher, especially in commodities sectors. In contrast, hydrogen power would be applied in the transportation and chemical industries. The two technologies have different application scenarios but should develop at the same pace.

Carbon capture is more suitable for industrial sectors. Electrification is limited in some industrial sectors, including chemistry, metal and non-metal smelting, papermaking as they require fuel burning which cannot be substituted by the electricity. Adopting hydrogen power would require a new production process. In contrast, we believe carbon capture technology is more convenient in these instances. It can integrate with existing technologies and requires some extra equipment.

Hydrogen power is more suitable for transportation and chemical industries. In some sectors, it is the energy consumption model rather than costs that limit electrification. For example, electric power cannot support long-term transport such as aviation and shipping. But hydrogen power’s high energy density makes it the most feasible to replace the fossil energy.

6.4.3 Other Energy Resources Adopted Carbon Capture and Hydrogen Power to Meet 22% and 8% Energy Demand in 2060

6.4.3.1 Hydrogen Power

Given the carbon emission forecasts for major emissions contributors, we expect hydrogen power consumption volume will rise to 110mnt (540mnt of standard coal) and 130mnt (650mnt of standard coal) in 2060 in neutral and optimistic scenarios under carbon neutrality, fulfilling 8%–10% of energy demand in 2060.

Transportation: With hydrogen power’s high mass-energy density and low volumetric energy density, hydrogen fuel cells can be used in long-distance transport scenarios without strict space requirements, such as highways, heavy-duty trucks, and aviation. Its fast charging feature also facilitates commercialization. In the optimistic scenario, the CICC transportation team expects 100% hydrogen-fueled heavy-duty trucks and the use of hydrogen power in aviation and shipping sectors.

Chemical industry: We expect to see hydrogen power in synthetic ammonia and methanol production. We think synthetic ammonia production can transform from alkaline electrolysis cells (AEC) to proton exchange membrane electrolysis (PEM) and finally to solid oxide electrolysis cells (SOEC), achieving 100% replacement of fossil fuels by hydrogen power. Methanol production will likely see similar transitions.

Manufacturing: We expect manufacturing subsectors to raise their concentration ratios via merger and acquisition, as well as market competition. As the total number of players reduces, advanced processes can be more widely applied.

6.4.3.2 Carbon Capture

Given our technology path forecast, some raw materials or fuels cannot be replaced by electric or hydrogen power. Based on estimation by CICC sector research teams, around 22% of fossil-fuel energy (equivalent to 1.34bnt of standard coal) cannot be replaced, which accounts for around 3.4bnt in carbon dioxide emissions.

Cement: According to the estimation from Gao Changming and other academics, they expect that the theoretical substitution ratio (TSR) in China’s cement industry to reach 25% in 2030 and 70% in 2050. Carbon dioxide emissions from calcium carbonate consumption and fuel consumption are 0.36 tonnes and 0.18 tonnes per tonne of cement. Even considering the reduction in the total output of cement through recycling and overall infrastructure development, the remaining carbon dioxide will still need to be offset by carbon capture.

Steel: If directly reducing iron ore in a blast furnace, it is possible to use 100% non-fossil fuels but not feasible for production processes. So the CICC commodity team’s estimate is made on the assumption that steel production requires fossil fuels and that coke remains a vital reducing agent.

Aluminum: Aluminum oxide production needs carbon capture technology as it cannot avoid the use of fossil fuels.

6.4.4 Costs and Green Premiums of Non-power Carbon Neutrality Solutions

As mentioned earlier, we estimate that electric power will not meet the remaining 30% energy consumption demand in our carbon neutrality case. To achieve zero-emissions in non-power energy consumption, we estimate 33% of energy can be replaced by hydrogen power, and the rest, i.e. coal, petroleum, and natural gas will need to adopt carbon capture technology. Based on these assumptions, we compare the carbon neutrality and base cases, and estimate the green premium of the non-power energy supply at 175% in 2021. If the costs of fossil-fuel energy remain unchanged and that of carbon capture and hydrogen power decline, we expect the sector green premiums of non-power sector to fall to 57% in 2030 and 14% in 2060, still above zero. In sum, to achieve carbon neutrality, the overall cost of non-power energy consumption will likely rise.

For hydrogen power, we estimate the per unit hydrogen-based power costs will fall from around Rmb13,000/tonne of standard coal in 2021 to around Rmb4,000/tonne of standard coal in 2060, still higher than the estimated weighted average power costs of around Rmb2,400/tonne of standard coal in 2021. Thus, hydrogen power will still have a green premium by 2060.

For carbon capture, we expect that per tonne carbon dioxide capture costs to fall from Rmb465/tonne (our estimate) in 2021 to Rmb262/tonne in 2060. As carbon capture is applied in the end-process, it creates extra costs compared with the base scenario and its green premium will stay above zero.

Based on our estimates, we note the cost of hydrogen power is lower than the diesel oil plus carbon capture solution (Fig. 6.10). As it is difficult for coal power or natural gas to adopt carbon capture technology, it would be easier to use hydrogen power in diesel-based consumption scenarios. Given practices in foreign countries and subsidy policy for renewable energy in China, we suggest that supporting policies kick in when the green premium of non-electricity energy technologies, i.e., hydrogen power and carbon capture, is near 200%.

Fig. 6.10
figure 10

Source CICC Research

Comparison of non-power zero-carbon solutions.

6.5 Policy Recommendation: Power System Reform Accelerates Non-fossil Energy Absorption and Helps the Development of Hydrogen Power

6.5.1 Policy Recommendation for the Power Sector: Stabilize New Energy Absorption

First, continue to improve the ancillary power system service market. Improving ancillary power system services is a good choice for China, as power storage is still not economically feasible. The government can enhance pricing mechanisms, incentivizing flexible power sources (thermal power with adjustable output, pumped-storage hydropower, power storage and demand response) to participate in power dispatching. Thus, when renewable resources account for the majority of output, it can lower the proportion of flexible resources to reduce the waste of renewable electric power.

Second, begin to adopt emission trading system. We think the government should build support mechanisms for emission trading system. Electricity’s spot prices reflect real-time demand and supply conditions. This facilitates peak shaving and valley filling. Moreover, it enables peak-valley spread arbitrage, accelerating the growth of energy storage demand. However, a mature trading system take time and effort to build. The spot market may face uncontrollable risks, such as fluctuations in prices. When electricity spot market can guide flexible resource investment, as well as operation and demand response, we expect the government to build effective supporting mechanisms to avoid potential risk in extreme scenarios.

Third, promote pilot energy storage projects. Energy storage helps enhance the stability and flexibility of new energy power output. We believe this is an effective way to raise renewable energy absorption capability in the parity era of “new energy plus energy storage solutions”. We believe energy storage reform at new-energy power plants is the most effective and least challenging way to promote energy storage. It doesn’t need a new market mechanism and will not challenge the power grid’s safety requirements. However, such a reform may lower ROI. Apart from policy support, we think the government can launch pilot energy storage projects, offering subsidies, tax reductions, and preferential policies to facilitate project developers’ land use and grid connections.

6.5.2 Policy Recommendation for Non-power Sector: Building Supporting System for Carbon Reduction; Establishing Reward and Penalty System to Fuel Clean Energy Development

First, the government may build supporting systems for carbon reduction. Taking coal-to-gas and coal-to-electricity programs as an example, they may encounter problems including unstable natural gas supply, subsidy delays, and insufficient infrastructure construction such as transmission pipelines. Apart from technology availability, we think the supporting system is crucial in planning low-carbon and zero-emission technology promotion.

Second, the government should launch a carbon trading rewards and punishment system to make the industrialization of new technology more economical. We expect carbon trading to accelerate upgrades of low-carbon technology in the non-power sector by limiting total emissions. Companies with large emissions will suffer environmental costs that weigh on production and operation and weaken their competitiveness. We expect carbon trading to push companies to reduce carbon emissions so as to lower costs. We believe subsidies for zero-emission or carbon-negative technologies (e.g., hydrogen power and carbon capture) will likely motivate companies to make the required changes to their operations.