In February of this year, China’s annual statistical communiqué indicated a change to an important measure of climate action.
The scope of carbon intensity, meaning the amount of carbon emitted per unit of economic output, now included CO2 emissions from industrial processes like cement-making. Before, it had only covered emissions from burning fossil fuels to generate energy plus using them as feedstocks.
Using the new scope, the communiqué concluded that China has achieved “17.7% carbon-intensity reduction”, slightly below the 18% goal set under the 14th Five Year Plan (2021-2025).
However, using the old scope the drop would have been 13-15%, we estimated at the Institute for Global Decarbonization Progress. China’s change to how it measures domestic carbon intensity has produced an understandable question about motivation and impact – is this a technical improvement, or a weakening of ambition?
A recent commentary argued that the change “effectively halved the apparent increase in China’s CO2 emissions over the past five years” and would make the “climate pledge easier to meet”.
The criticism identifies a genuine problem with the frequency and extent of information disclosure, but it overstates what the accounting change does.
In reality, this change indicates a scope expansion of China’s carbon emissions sources covered by the binding targets in the 15th Five Year Plan (2026-2030). It prepares China’s shift from an intensity-only approach to a “dual carbon mechanism” that combines intensity reduction with an absolute emissions cap.
As part of its latest climate action plan under the Paris Agreement – known as its Nationally Determined Contribution, or NDC – China has pledged to reduce CO2 emissions per unit of GDP by more than 65% from 2005 levels.
Though the broadened five-year plan indicator is important for domestic implementation, it cannot by itself alter the international base year, the NDC scope or the final 2030 assessment.
As we will see, the scope changenarrows the gap between domestic and international indicators, but it does not rewrite the NDC. The useful question is therefore not whether the old boundary was inconvenient, but whether the new one is more complete, more durable and more transparent?
Changed and unchanged
China uses more than one carbon-accounting framework.
Internationally, as a party to the Paris Agreement, China is required to disclose emissions data and progress toward its NDCs through documents such as biennial transparency reports or national communications.
In such documents, China evaluates its progress on implementing NDC goals, including carbon-intensity reduction. These reports are based on the national greenhouse gas emissions inventory which is released roughly every couple of years. The inventory covers all emission sources as well as carbon removal, in accordance with the UN climate convention’s guidelines.
Domestically, carbon-intensity targets in five-year plans are implementation tools: they guide national, provincial and sectoral performance assessments on a much faster cycle than NDCs. Their goals are set on tighter timescales, and their results usually summarised in annual statistical communiqués.
The confusion comes from the fact that these two systems have not yet been fully aligned in practice. Carbon intensity in the five-year plan has historically focused on energy-related CO2 from fuel combustion. That covered above 80% of the emissions and made the indicator easier to update annually. But it deprioritised the industrial sources outside the binding domestic performance framework.
What has driven China’s carbon-intensity scope change?
One important motivation is to more comprehensively manage carbon. Incorporating process emissions not only brings domestic accounting closer to international practice, but also creates stronger incentives to improve emissions accounting, along with statistical capacity at provincial and local level.
The new domestic boundary brings industrial process emissions into a binding carbon-intensity obligation. Rather than focussing only on emissions from energy burned as fuel, process emissions from cement-clinker manufacturing, iron and steel, chemical production or other feedstocks are now included.
These sources cannot be reduced by only electrifying equipment or switching to a zero-emission power grid. They require clinker substitution, output reduction or carbon capture. Bringing these hard-to-abate sources under a mandatory intensity target is a step forward in governance. Local governments and enterprises now face a harder constraint on sources that previously sat outside the main five-year-plan carbon-intensity assessment.
The coal-to-chemicals industry, through which solid coal is turned into products like methanol, ammonia and olefin, deserves a brief note. In the coming years the industry may produce more emissions. This growth could be driven by demand for chemical products, the costs of competing fossil-fuel feedstocks, regulations on burning coal for fuel, and concerns about supply-chain resilience.
Including it in the expanded boundary is therefore positive. It brings a rising source under domestic carbon-intensity discipline. At the same time, coal-to-chemical emissions remain much smaller than China’s largest non-power-related sources, namely process emissions from cement, iron and steel. And their growth may not continue indefinitely as demand, costs, policy controls and substitution options evolve.
It is fair to acknowledge that the timing also makes the 14th Five Year Plan (2021-2025) result look closer to the target, especially during a sharp decline in cement production, driven by the real-estate downturn. But from a regulatory perspective, incorporating a sector during a period of structural decline is entirely logical. It embeds the reduction in future policy baselines and subjects any subsequent rebound in activity to carbon-intensity constraints.
As the international-domestic gap continues to narrow, we can reasonably expect that five-year-plans will very likely cover all greenhouse gases, given that China has committed cutting all such emissions by 7-10% from peak level by 2035.
Does the change make the 2030 target easier?
In an accounting sense, no. Progress toward the 2030 pledge is evaluated through national inventory data and international transparency reporting, not by retrospectively changing one annual statistical communiqué. As mentioned, the five-year-plan indicator cannot by itself alter the NDC’s scope or the final assessment in 2030.
It is also risky to infer China’s actual emissions trajectory by back-calculating from aggregate carbon-intensity figures alone. While such exercises can provide useful illustrations, they are subject to significant uncertainty. Carbon intensity depends on GDP data, which are periodically revised through economic censuses, statistical adjustments and updates to account for inflation.
Process emissions are not published through the same routine statistical channels as energy-related emissions. Analysts therefore have to use assumptions, proxies and reconstruction methods. Such work can be valuable for scenario analysis, but it cannot fully replicate official inventory accounting.
For the same reason, the scope change is not directly relevant to China’s CO2 peaking target. Peaking is about total CO2 emissions as measured in inventory terms, not about a five-year domestic intensity indicator. While a more thorough domestic assessment metric can help bring policy pressure to harder-to-abate emissions sources, it does not move the peak date or erase emissions from the inventory.
Transparency and frequency matter
The real lesson from this debate is that China needs more transparent and timely climate data. It is true that, under the Paris Agreement framework, it does not need to publish an annual greenhouse gas inventory. But researchers and journalists naturally rely on more frequent data and proxies when official data is not available; the gap between annual domestic indicators and less frequent inventory-based reporting leaves room for misinterpretation.
Trust matters, from both sides. Near-real-time monitoring estimates are valuable for trend signals, but they are not designed for definitive accounting. When a proxy estimate and an official inventory figure diverge, it is more productive to start by asking what the proxy misses. Jumping straight to consider what the official figure has concealed does not advance transparency. It undermines the credibility of the analysis and erodes the mutual trust that international climate cooperation needs.
Transparency matters. It is important to call for more frequent, more granular greenhouse gas inventory data – broken down by sector, subsector and source category. This will mean future analyses do not need to rely on proxies that conflate combustion and process emissions, aggregate across sectors, and fail to capture structural shifts in fuel end-use.
