Blog

China and India Drop Coal

Jan Gunash · Co-Founder & CGO · May 18, 2026

The best questions are not always the comfortable ones. Like: when the headline says the energy transition is working, what exactly is being measured? And when two of the most consequential data points in the history of climate change arrive in the same month and barely register in the news cycle, what does that say about the metrics the world is using to track its own progress?

These are not rhetorical questions. Two things happened in the global energy system in 2025 that the data had been pointing toward for years. Neither was inevitable. Both were the result of decisions made years earlier, in boardrooms and planning ministries, on the basis of data that the people making those decisions trusted. This newsletter is about what the data showed, what it almost missed, and what it still cannot see.

The First Time in 52 Years

In 2025, coal power generation fell in both China and India simultaneously for the first time since 1973. That year was the oil crisis. The reason coal fell then was a global demand shock that nobody planned for. The reason it fell in 2025 is the opposite: deliberate, sustained investment in clean energy at a scale that finally exceeded demand growth (source: Carbon Brief).

The numbers are worth sitting with. Coal fell by 1.6% in China, equivalent to 58 terawatt-hours, and by 3.0% in India, equivalent to 57 terawatt-hours. Those are not large percentage drops. They are historically significant ones. The power sectors of these two countries drove 93% of the rise in global carbon dioxide emissions from 2015 to 2024 (source: Carbon Brief). When the direction of travel changes in both of them at once, the implications for global emissions trajectories are significant.

What drove the drop is as important as the drop itself. In 2025, China added more than 300 gigawatts of solar and 100 gigawatts of wind power, both clear records for any country ever. Power generation from solar and wind increased by 450 terawatt-hours (source: Carbon Brief). India followed a similar path, adding 35 gigawatts of solar, 6 gigawatts of wind, and 3.5 gigawatts of hydropower in the first 11 months of 2025, with renewable capacity additions up 44% year on year (source: Carbon Brief / Eco-Business).

Faster clean-energy growth contributed 44% of the reduction in coal and gas use compared to the trend from 2019 to 2024, while milder weather contributed 36% and slower underlying demand growth contributed 20% (source: Carbon Brief). Clean energy was the largest single driver. Not luck. Not a demand slump. Investment.

Both countries now have the preconditions in place for peaking coal-fired power, if China is able to sustain clean-energy growth and India meets its renewable energy targets (source: Carbon Brief). That is not a guarantee. It is a trajectory. And it is the first time the data has supported that sentence being written honestly rather than aspirationally.

What the Metrics Almost Missed

Here is the data problem hiding inside the good news. The metrics most widely used to track the energy transition were not designed to surface this moment. Annual percentage changes in coal generation are reported with a multi-month lag. The data that would allow a real-time picture of whether clean energy is outpacing demand growth in the world's two largest coal consumers does not exist in a publicly accessible, standardised form. Carbon Brief's analysis is the work of a single analyst synthesising data from multiple national energy agencies that report in different formats, at different frequencies, with different methodologies.

The milestone happened. The data infrastructure to make it visible as it happened did not exist. That gap matters because the decisions that determine whether this trajectory is sustained, the policy commitments, the investment signals, the grid planning assumptions, are being made by people who depend on the same data systems that almost missed the milestone they are being asked to build on.

The Threshold Nobody Was Watching

The second story happened in Europe. Wind and solar generated more electricity across the European Union than fossil fuels for the first time in 2025, supplying a record 30% of EU power against fossil fuels at 29% (source: Ember / Electrek).

Beatrice Petrovich, lead analyst at Ember and author of the report, described it as a major tipping point of strategic importance to the EU. The danger of relying on fossil fuels looms large in destabilised geopolitics, she noted, and the Hormuz crisis that ran through this newsletter's previous four editions makes that observation concrete in a way that a January energy review could not have anticipated (source: Ember / Canada's National Observer).

Solar led the growth, with generation jumping by more than 20% for the fourth consecutive year, reaching 369 terawatt-hours in 2025. Coal fell to a historic low of 9.2% of EU power, with 19 EU countries now below 5% coal in their electricity mix (source: Euronews / Ember). In 14 of the EU's 27 member states, wind and solar generated more electricity than all fossil fuels combined. Five years ago that number was three countries. The structural shift is not evenly distributed but it is accelerating (source: Euronews).

The Data Gap Inside the Good News

The EU story has a data infrastructure caveat the headlines did not carry. Congestion management costs in Europe neared 9 billion euros in 2024, while 72 terawatt-hours of mainly renewable energy was curtailed, shut off due to grid bottlenecks. That is roughly the equivalent of Austria's entire annual electricity consumption, generated and then wasted because the grid was not designed to move it (source: Euronews / Aurora Energy Research).

The data systems that track renewable generation are reasonably good. The data systems that track grid capacity, congestion costs, and curtailment in real time are not. A transition that produces more clean electricity than the grid can carry is a transition that is building faster than the infrastructure designed to measure and manage it.

What the Two Stories Have in Common

China and India cutting coal for the first time in 52 years. The EU crossing the renewable threshold for the first time ever. These are two of the most significant data points in the history of the energy transition and both of them arrived without the monitoring infrastructure to make them legible in real time.

The Carbon Brief analysis was published weeks after the year ended, synthesised by hand from national data sources that do not speak to each other. The Ember analysis aggregated across 27 national reporting systems with different methodologies and different lags. Neither finding was wrong. Both were produced by serious analysts doing serious work. The point is that the data infrastructure that should make milestones like these visible as they develop, rather than retrospectively, does not yet exist.

This is the version of the data gap argument that is hardest to make and most important. The previous editions of this newsletter covered gaps that allowed bad things to happen invisibly. The fertilizer cascade. The insurance model failure. The water bankruptcy. This edition is about a gap that is allowing good things to happen invisibly too. Progress that cannot be measured in real time cannot be reliably replicated, scaled, or defended when the policy environment turns against it.

The Path Forward

The data from China, India, and the EU in 2025 shows that the physical transition is moving faster than most of the models predicted. The question is whether the data infrastructure, the governance frameworks, and the policy commitments can catch up with the physical reality that is already underway.

The metrics used to track the energy transition were designed for a different phase of it. They measure installed capacity, annual generation, and percentage shares. They are not designed to show whether clean energy is outpacing demand growth in real time, whether grid infrastructure is keeping pace with generation, or whether the preconditions for peaking coal are being sustained across consecutive quarters. Building those metrics is not a technical problem. It is a governance and investment problem. The data exists. The frameworks to make it actionable do not.

Two countries accounting for most of the world's emissions growth cut coal simultaneously. The continent most exposed to energy geopolitics crossed the renewable threshold. The transition is working. The systems designed to tell us how fast, how reliably, and what comes next are not keeping pace. That is the next problem worth solving.

Sources:

Want to contribute your thoughts or findings in data and AI with climate, energy, or nature? Message Jan Gunash on LinkedIn or send an email to jgunash@projectcoeus.com.